(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "soil analysis"

SOIL CHEMISTRY 

THIRD EDITION 

HINRICH L BOHN 

BRIAN L. McNEAL 

GEORGE A. O'CONNOR 



® 



JOHN WILEY & SONS, INC. 

New York / Chichester / Weinheim / Brisbane / Toronto / Singapore 



This book is printed on acid- free paper. @ 

Copyright © 2001 by John Wiley & Sons, Inc. All rights reserved. 

Published simultaneously in Canada. 

No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or 
by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except as 
permitted under Sections 107 or 108 of the 1976 United States Copyright Act, without either the prior 
written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to 
the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 
750-4744. Requests to the Publisher for permission should be addressed to the Permissions Department, 
John Wiley & Sons, Inc., 605 Third Avenue, New York, NY 10158-0012, (212) 850-601 1, fax (212) 
850-6008. E-Mail: PERMREQ@WILEY.COM. 

This publication is designed to provide accurate and authoritative information in regard to the subject 
matter covered. It is sold with the understanding that the publisher is not engaged in rendering 
professional services. IT professional advice or other expert assistance is required, the services of a 
competent professional person should be sought. 

Library of Congress Cataloging-in-Publkation Data: 
Bohn, Hinrich L., 1934- 

Soil chemistry / Hinrich L. Bohn, Brian L. McNcal, George A. O'Connor. — 3rd ed. 

p. cm. 
Includes bibliographical references (p. ). 
ISBN 0-47 1 -36339-1 (cloth : alk. paper) 

1. Soil chemistry. I. McNcal, Brian Lester, 1938- II. O'Connor, George A., 1944- III. 
Title. 

S592.5 .B63 2001 
631.4'1— dc2l 2001017914 

Printed in the United States of America 

10 9876 5 432 



CONTENTS 



Preface xi 

1 Introduction 1 

1.1 The Soil Solution / 2 

1.2 Background / 5 

1.3 Soil-Ion Interactions / 7 

1 .4 Colloids and the Soil Solution / 7 

1 .5 Compositions of Soils and Plants / 8 

1.6 Nonagricultural Soil Chemistry / 10 

1.7 Biogeochemical Cycles and Pollution / II 

1.8 Soil and the Hydrosphere / 12 

1.9 Soil and the Atmosphere / 14 

1.10 Soils and the Development of Life / 1 5 

1.11 The Role of Soil in the Environment 
and the Maintenance of Life / 17 

1.12 Chemical Units / 19 

2 Important Ions 26 

2. 1 Essential Elements / 30 

2.2 Toxicity and Deficiency / 30 

2.3 Alkali and Alkaline Earth Cations / 35 

2.3.1 Calcium / 37 

2.3.2 Magnesium / 38 



VI CONTENTS 

2.3.3 Potassium (Kalium) / 38 

2.3.4 Sodium (Natrium) / 39 

2.4 Major Soluble Anions / 40 

2.4.1 Halides / 42 

2.5 Weakly Soluble Anions / 44 

2.5.1 Boron, Silicon, Molybdenum, Arsenic, and Selenium / 44 

2.5.2 Phosphate / 47 

2.6 Aluminium, Hydrogen, and Transition Metals / 50 

2.7 Toxic Elements in Soils / 55 

2.8 Carbon, Nitrogen, and Sulfur / 61 

2.8.1 Nitrogen / 62 

2.8.2 Sulfur / 65 

3 Water and Solutions 68 

3.1 Acids and Bases / 71 

3.1.1 Hydrolysis and Deprotonation / 72 

3.1.2 Solubility Products / 75 

3.2 Chemical Activity / 76 

3.3 Complex Ions and Ion Pairs / 79 

3.4 Hard and Soft Lewis Acids and Bases / 84 

3.5 Soil Reaction Coefficients / 85 

3.6 Models of the Soil Solution / 87 

A3.1 Thermodynamics / 87 

A3. 1.1 Gibbs Free Energy / 90 

A3. 2 Solid Solutions and Open Systems / 93 

A3. 3 Kinetics / 97 

A3.3.1 Reaction Order and Rate Constants / 99 

A3.3.2 Temperature Effects / 101 

A3.3.3 Microbially Catalyzed Reactions / 102 

4 Oxidation and Reduction 107 

4.1 Soil Oxidation-Reduction / 109 

4.2 Electron Donors / 110 

4.3 Electron Acceptors /111 



CONTENTS Vli 

4.4 Redox Reactions / 112 

4.5 Flooded Soils / 115 

A4.1 Electrochemistry / 116 

A4.2 E/iandpe / 118 

A4.2.1 Nitrogen / 118 
A4.2.2 Sulfur / 122 
A4.2.3 Iron / 124 

A4.3 Redox Potential Measurements / 125 

5 Inorganic Solid Phase 129 

5.1 Crystal Chemistry of Silicates / 130 

5.2 Structural Classification of Silicates / 135 

5.2.1 Layer Silicates / 135 

5.2.2 Relation of Structure to Physical and 
Chemical Properties / 138 

5.3 Soil Layer Silicates / 140 

5.3.1 Kaolins / 140 

5.3.2 Smectites (Montmorillonite) / 141 

5.3.3 Vermiculites / 142 

5.3.4 Micas / 143 

5.4 Chlorites / 143 

5.5 Accessory Minerals / 144 

5.5.1 Allophane and Imogolite / 144 

5.5.2 Zeolites / 145 

5.5.3 Al, Fe, Ti, and Mn Hydroxyoxides / 145 

5.6 Charge Development in Soils / 146 

5.6.1 Permanent Charge / 146 

5.6.2 pH-Dependent Charge / 147 

A5.1 Surface Area Measurements / 151 
A5.2 Mineral Identification in Soils / 152 

6 Soil Organic Matter 155 

6. 1 Soil Organic Matter Content / 155 

6.2 The Decay Process / 158 



VIII CONTENTS 

6.3 Extraction, Fractionation, and Composition / 161 

6.4 Colloidal Properties / 164 

6.5 Function of Organic Matter in Soil / 167 

6.5.1 Organic Chemical Adsorption / 170 

7 Weathering and Soil Development 172 

7.1 Stability of Parent Material Minerals / 181 

7.2 Ionic Potential / 183 

7.3 Rates of Weathering and Soil Development / 185 

7.3.1 Acidity / 187 

7.3.2 Mechanisms of Mineral Decomposition / 188 

7.3.3 Time Sequence of Mineral Occurrence / 189 

7.4 Mineral Formation in Soils / 192 

7.4.1 Soil Carbonates / 197 

7.4.2 Carbon Dioxide / 198 

7.4.3 Evaporites / 200 

A7.1 Stability Diagrams / 201 

8 Cation Retention (Exchange) in Soils 206 

8.1 Electrostatic Cation Retention (Cation Exchange) / 207 

8.1.1 Exchange Selectivity / 212 

8.1.2 Cation Exchange Equations / 215 

8.1.3 Diffuse Double Layer / 2 1 7 

8.2 Strongly-Retained Cations / 221 

8.2.1 Oxide-Retained Cations / 221 

8.2.2 Cations Retained by Soil Organic Matter / 224 

A8.1 Diffuse Double-Layer Theory / 225 

A8.2 Cation Exchange Equations / 229 

A8.3 Determination of Cation Exchange Capacity 
and Exchangeable Cations / 233 

9 Anion and Molecular Retention 237 

9.1 Anion Exchange / 241 

9.2 Strong Anion Retention / 243 

9.2.1 Phosphate Reactions in Soils / 246 



CONTENTS IX 

9.3 Molecular Retention / 250 

9.4 Adsorption Isotherms / 254 

A9. 1 Multisite and Multilayer Adsorption / 256 

10 Acid Soils 260 

10. 1 Instability of Hydrogen Soils and Clays / 262 

10.2 Hydrolyzed Aluminium Ions / 263 

10.3 Classification and Determination of Soil Acidity / 265 

10.4 Soil pH Measurements / 268 

1 0.5 Percent Base Saturation / 268 

10.6 Lime Requirement / 269 

10.7 Aluminium and Manganese Toxicity / 272 

10.8 pH and Macronutrients / 272 

10.9 pH and Micronulrients / 273 
10.10 Management / 274 

A 10.1 pH and Ion Activity Measurements / 275 

11 Salt-Affected Soils 280 

11.1 Distribution and Origin / 280 

11.2 Irrigation Water Quality / 284 

11.2.1 Sodium Hazard / 286 

11.2.2 Bicarbonate Hazard / 288 

1 1.2.3 Other Toxic Solutes / 289 

11.3 Characterizing Salt-Affected Soils / 289 

1 1 .4 Effects of Salts on Soils and Plants / 29 1 

1 1.5 Salt Balance and the Leaching Requirement / 294 

11.6 Reclamation / 298 

A 11 . 1 The Langelier Index / 300 

Index 303 



PREFACE 



I thank Linda Candelaria, Gavin Gillman, Robert Hatter, Mark Noll, Stom Ohno, and 
Scott Young for their suggestions and encouragement to write a third edition of Soil 
Chemistry. I am especially grateful to Linda Candelaria for her excellent suggestions 
on additions and revisions. 

This edition tries to emphasize that the soil and soil solution are the center and 
heart of the environment. The chemical composition of the biosphere, hydrosphere, 
and atmosphere depends greatly on the chemistry of the soil. 

Since the second edition appeared, much of the interest in soil chemistry has been 
on the fate of so-called toxic chemicals and elements in soils. This edition points 
out that (1) all of the chemical elements — toxic and beneficial — were always in the 
soil, (2) the soil is the safest part of the environment in which to deposit our wastes, 
(3) there are wise and unwise ways to utilize soil for waste disposal, (4) soil chem- 
istry degrades wastes and converts them into benign or useful substances, (5) envi- 
ronmental activists and the popular media usually ignore the dose-response concept 
that is central to toxicology and to soil fertility, and (6) how much is in the soil, how 
fast it is changing, and how easily it transfers to plants and water are more important 
than what is there. Soil chemistry can answer those important questions. A goal for 
the future is to answer them better. 

The use of aluminium for aluminum, occasionally of natrium and kalium for 
sodium and potassium, and often of essential elements for essential nutrients is not 
affectation. The intent is to bring more international usage and correctness into the 
book. Soil chemistry is a global issue. 

Brian McNeal and George O'Connor were unable to work on this edition of our 
book. 1 retained much of their excellent contributions to the previous editions. Any 
errors or omissions are my responsibility. I would be grateful if readers would call 
them to my attention. 

HlNRICH BOHN 

Tucson, Arizona 



XI 



1 



INTRODUCTION 



No one regards what is at his feet; we all gaze at the stars. 

— Quintus Ennius (239-169 BC) 

Heaven is beneath our feet as well as above our heads. 

—Henry David Thoreau (1817-1862) 

The earth was made so various that the mind of desultory man, studious 
of change and pleased with novelty, might be indulged. 

—William Cowper (The Task, 1780) 

The quotations illustrate how differently humans see the soil that gives them life and 
feeds them. Those opinions have been held for a long time. Most people are still at 
the knowledge level of Quintus Ennius who lived more than 2000 years ago. They 
take for granted the food that the soil produces, the clean water and air that the soil 
provides. Thoreau's and Cowper's wonder and fascination of soils is rarely expressed 
or felt. Yet the soil is wondrous if one looks closely. The soil — the solid but porous 
surface of the earth to about one meter depth, the depth that roots penetrate — has 
many mysteries. The soil is as mysterious and exciting as any other science and any 
other part of the universe. 

Soil is a mixture of inorganic and organic solids, air, water, microorganisms, and 
plant roots. All these phases influence each other: Weathering and adsorption by the 
soil affect air and water quality, air and water weather the soil, microorganisms cat- 
alyze many of the reactions, and plant roots absorb and exude inorganic and organic 
substances. Soil chemistry considers all these reactions but emphasizes the reactions 
of the soil solution, the thin film of water and its solutes (dissolved substances) on 
the surfaces of soil particles. 

1 



2 INTRODUCTION 

1 .1 THE SOIL SOLUTION 

The soil solution is the interface between soil and the other three active environ- 
mental compartments — atmosphere, biosphere, and hydrosphere (Fig. 1.1). The 
boundaries are dashed lines to indicate that matter and energy move actively from 
one compartment to another; the environmental compartments are closely inter- 
active rather than isolated. The interface between marine sediments and seawater, 
and between groundwater and subsoils, is chemically much the same as the interface 
between surface soils and the soil solution. Sediments remove and release ions from 
the bodies of water they contact by the same processes as the interface between the 
soil and the soil solution. 

The soil solution is the source of mineral nutrients for all terrestrial organisms. As 
the soil solution percolates below the root zone, it becomes groundwater or drains to 
streams, lakes, and the oceans, and strongly affects their chemistry. The amounts of 
matter transferred are much greater and the rates of these reactions are much faster in 
the soil than in the other environmental compartments. The soil solution is the most 
important transfer medium for the chemical elements that are essential to life. 

The soil solution differs from other aqueous solutions in that it is not electrically 
neutral and usually contains more cations than anions. The net negative charge of 
soil clay particles in most soils extends electrically out into the soil solution, and the 
charge is balanced by an excess of cations in the soil solution. These cations belong 
to the solid but are present in the soil solution. Soils in old and heavily weathered 
soils, as in parts of Australia, Africa, and South America, or in soils of volcanic 
origin, as in Japan and New Zealand, may have a net positive charge. There the soil 
solution has an excess of anions. 

The interactions of ions and electrical charge at the soil particle-soil solution 
interface happens at all particle interfaces. In cases outside of the soil, this interaction 



Atmosphere I Biosphere 



^ / Soil Solution \ 

i / \ *\ 

Soil Solids ] I 



r 

Hydrosphere 
FIGURE 1.1. The soil and the soil solution are the heart of the environment. 



THE SOIL SOLUTION 3 

is generally negligibly small. The soil is unique because the soil's surface area is 
so large that this interaction becomes so extensive. Because of this interaction, the 
boundary between soil solids and the soil solution is diffuse. The water and ions at 
the interface belong to both the aqueous phase and to the soil solids. 

The diffuse layer extends out as much as 50 nm into the aqueous solution from the 
particle surface. For clay (colloidal)-size (<2 ^.m) particles with their large surface 
area, this interaction is great enough to significantly affect the composition of the 
soil solution next to the colloidal particles. Because soils contain considerable clay, 
a large part of the soil solution is affected by colloids. At the so-called field capacity 
water content, most of the soil solution is in the < 10 jj.m contacts and pores be- 
tween sand (50-2000 fj.m) and silt (2-50 /nm) particles. Clay particles and microbes 
congregate at these contacts so the soil solution interacts closely with these reactive 
bodies in these contact zones. The soil solution on open sand and silt surfaces is only 
10-100 nm thick, so much of this water is also affected by the particle's charge. The 
portion of soil solution affected by soil colloids increases as the soil dries. 

Most soil reactions occur at the soil solution/soil interface. Ions in water can move 
and react fast enough to measure easily. Slower but still measurable reactions occur 
in the weathered surfaces of soil particles. These poorly understood surfaces con- 
tain considerable water. Reaction rates in the truly solid phase at soil temperatures, 
however, are too slow to be measured in our lifetimes. 

Because the mass and reactivity of soils are great, the chemistry of me atmosphere 
and fresh water are largely controlled by the chemistry of the soil solution. Reactions 
that require days and years in air, and hours in water, require only seconds and min- 
utes in soils. The compositions of the air, water, and biomass compartments in the 
environment evolved from, and still respond to, the chemistry of the soil. The soil 
came first and as it changed, it changed the others. The change in the others also 
changed the soil, but to a lesser extent because of the soil's mass. 

The soil solution contains a wide variety of solutes, including probably every 
element in the periodic table. This book discusses those solutes that are active in the 
environment — the solutes that affect plant and animal life, or have been of concern 
in pollution. 

Figure 2.1 shows the chemical elements that are essential to living organisms, 
those that are reactive, and those that are in significant amounts in each environmen- 
tal compartment. Chapter 2 discusses the life-essential and other important elements 
in more detail. All of the major and minor essential elements are stored and available 
for transfer in soils; the amounts in the other compartments, although important, are 
generally much, much smaller. Major and minor refer only to the amounts needed by 
organisms; all are essential and all are needed in their proper amounts. Too little is 
deficiency, too much is toxicity. The essential elements are also called essential nu- 
trients, but elements is a better term because nutrients implies energy content, such 
as in the carbohydrates and fats of food. 

With few exceptions, the soil supplies these elements to living organisms in about 
the right amounts. Although not perfect, the soil's supplying power is remarkably 
effective. This is not simply a happy coincidence; life evolved in response to the 
availability of these elements in soils. 



4 INTRODUCTION 

For most people the major reason to study soil chemistry is to insure and to in- 
crease production of food and fiber crops. The soil is and will be the main source 
of human nutrition. The oceans supplement our food supply but their productivity 
is limited by the osmotic potential of the water, the limited availability of essen- 
tial elements, the low temperature of ocean water, and the long food chain between 
photosynthetic organisms and those large enough to harvest. The oceans can pro- 
vide only a small amount of society's needs and wants. The soil's productivity per 
unit area is many times that of the oceans. Terrestrial plants remain the cheapest and 
best means of converting solar energy into life support for this planet. The growth 
of plants on soils is the basis of most of the world's economy and of a nation's well 
being. 

Soil chemistry is only one of many factors that affect plant growth. In contrast 
to climate and other uncontrollable factors, however, agriculturalists can influence 
and modify soil chemistry to considerable extent. The amounts of essential elements 
needed by plants over a season are small enough that supplementing the soil supply 
is feasible. Increasing the efficiency of that fertilization is a continuing soil chemistry 
challenge. The toxicity of materials that harm plant growth can also be controlled by 
soil chemistry. 

People are now learning to appreciate the soil's large role in the biogeochemical 
cycling of the elements. Soils can mitigate many undesirable human-caused changes 
(pollution) of the environment. Safe removal of wastes from the environment has 
been recognized to be as important for continued civilization as food production. 
The retention, exchange, oxidation, and precipitation of waste in soils make them 
unequaled as recycling media. 

In earlier times when the population was less dense and industries were few and 
small, wastes were distributed widely on and in the soil and could readily return to 
their natural biogeochemical cycles. By concentrating wastes in urban areas, indus- 
trial facilities, landfills, feedlots, and sewer outfalls — releasing wastes to the air and 
water rather than allowing them to react in soils — by fertilization, and by creating 
synthetic chemicals that react slowly, humanity has occasionally and in local areas 
exceeded the rate at which these materials can return to their biogeochemical cycles. 
"Advanced" societies sometimes overlook the degradative functions of soil and look 
instead for expensive, and only partially satisfactory, technological methods of waste 
disposal. Humanity creates pollution, which has awakened a new awareness of the 
importance of soil chemistry. 

Soil chemistry is closely related to colloid (surface) chemistry, geochemistry, soil 
fertility, soil mineralogy, and soil microbiology or biochemistry. Soil fertility con- 
siders soil as a medium for plant growth. Soil mineralogy examines the structural 
chemistry of the solid phase. Soil microbiology studies soil biochemical reactions. 
Such subdivision is necessary to study the soil thoroughly, but these subdivisions 
sometimes obscure the interaction between soil components, and this interaction is 
often as important as the properties of the components alone. 

Soil chemistry traditionally has had two branches: inorganic and organic, but strict 
separation of the two fields is difficult and pointless in many cases. The direction 
of biochemical soil reactions is largely based on the inorganic phase. Soil organic 



BACKGROUND 5 

processes affect primarily the rate of soil chemical reactions. Biochemical reactions 
are carried out by soil microorganisms, whose vast numbers in soils influence many 
reactions. For several elements, notably carbon, nitrogen, and sulfur, the microbial 
role almost totally determines soil reaction rates. Biochemical and microbial reac- 
tions are primarily catalytic processes affected by the independent variables of soil 
mineral composition, climate, gas exchange with the atmosphere, and energy from 
photosynthesis. Despite the importance of biochemical reactions, research in soil 
chemistry historically has been more oriented toward inorganic processes. 



1.2 BACKGROUND 

Food and fiber production were already important before agriculture began. After 
fear, food is the dominant concern of every animal. The senate of ancient Athens de- 
bated soil productivity 2500 years ago and voiced the same worries about sustaining 
and increasing soil productivity that are heard today. Can this productivity continue 
or is soil productivity being exhausted? 

In 1790, Mallhus noticed that the human population was increasing exponentially 
and that food production was increasing arithmetically. He predicted that by 1850 the 
demand for food due to population growth would overtake food production, and peo- 
ple would be starving and fighting like rats for morsels of food. Similar apocalyptic 
predictions continue to crop up and cannot be disregarded. It is encouraging, how- 
ever, that productivity has increased since the Greek senate debates and faster than 
Malthus predicted. In recent history food productivity has been increasing faster than 
ever. The earth now feeds the largest human population ever and a larger fraction of 
that population is better fed than ever before. Whether this can continue and at what 
price to the environment and other organisms is an open question. One encouraging 
part of the answer is the rapidly declining human birth rate on most continents in 
recent decades, thus putting less stress on the soil's resources in the future. Another 
part of the answer lies in soil chemistry, and much progress is still to be made in our 
understanding of the soil and its potential. 

Agricultural practices that increase crop growth — planting legumes, manuring 
with animal dung and with litter from forests, rotating crops, and liming — were 
known to the Chinese 3000 years ago. These practices had also been learned by 
the Greeks and Romans and appeared in the writings of Varro, Cato, Columella, 
and Pliny. The reasons for their effectiveness, however, were unknown. Little or no 
further progress was made in the Western world for almost 1500 years because of ig- 
norance and deductive reasoning. Deduction is applying preconceived ideas, broad 
generalities, and accepted truths to particular problems, without testing if the pre- 
conceived ideas and accepted truths are valid. One accepted truth, derived from the 
Greeks, was that matter was composed of earth, air, fire, and water — a weak basis, 
as we later learned, on which to increase knowledge. 

In the early 1500s, Sir Francis Bacon pointed that inductive reasoning, the sci- 
entific method, is a much more productive approach to gaining new knowledge — 



6 INTRODUCTION 

observe and measure, derive broader ideas from the data, and test these ideas again. 
The scientific method brought progress, but the progress in soil chemistry was slow. 

Palissy (1563) thought that plant ash came from the soil and when added back 
to the soil could be reabsorbed by plants. Plat (1590) thought that salts from decom- 
posing organic matter dissolved in water and absorbed by plants were responsible for 
plant growth. Glauber (1650) thought that saltpeter (Na, K nitrates) was the key to 
plant nutrition by the soil. Kuelbel (ca. 1700) believed that humus was the principle 
of vegetation. Boerhoeve (ca. 1700) believed that plants absorbed the "juices of the 
earth." Others have found this idea in Pliny's writings. None of them, however, had 
experimental proof. 

Van Helmont (1592) tried to test these ideas. He planted a willow shoot in a pail 
of soil and covered the pail so that dust could not enter. He carefully measured the 
amount of water added. After five years the tree had gained 75.4 kilograms. The 
weight of soil in the pot was still "200 lb (90.8 kg) less about two ounces (56 g)." 
Van Helmont disregarded the 56 grams as what we would today call experimental 
error. He concluded that the soil contributed nothing to the nutrition of the plant and 
that plants needed only water for their sustenance. Although he followed Bacon's 
suggestions and the scientific method as well as he could, he unfortunately came to a 
wrong conclusion. Many experiments in nature still go afoul because of incomplete 
control and measurement of all the experimental variables. 

John Woodruff's (1699) experimental design was much better. He grew plants in 
rainwater, river water, sewage water, and in sewage water plus garden mould. The 
more solutes and solids in the growth medium — the "dirtier" the water — the better 
the plants grew, implying that something in soil improved plant growth. The idea 
developed, but without further testing, that the organic fraction of the soil supplied 
the plant's needs. That idea persists to this day. Organic substances absorbed by 
plants from the soil may affect plant growth, but this has been difficult to prove. 

In 1840 Justus von Liebig persuasively advanced the idea that inorganic chemi- 
cals were key to plant nutrition and that an input-output chemical budget should be 
maintained in the soil. Liebig's theory was most probably based on Carl Sprengel's 
work in 1820-1830 that showed that mineral salts, rather than humus or soil organic 
matter, were the source of plant growth. Liebig's influence was so strong that when 
Boussingault (1865) measured more nitrogen appearing in plants than he put into the 
soil, his work was disregarded for many years. Microbial nitrogen fixation did not fit 
into the Sprengel-Liebig model. 

Soil chemistry was first recognized as distinct from soil fertility in 1850 when 
Way and Lawes, at Rothamsted, England, discovered cation exchange. Their work 
suggested that soils could be studied apart from plants; yet the results would still 
have implications for soil fertility. 

The Information Age is upon us and the information is both accurate and inac- 
curate. From the information available about nutrition and health, for example, one 
might deduce that life is becoming riskier. In reality we are living longer and health- 
ier lives than ever before. To make sense of what we hear and read, we sometimes 
still resort to deductive reasoning. Science no longer has the certainty that it once 
seemed to have and can be very complicated. People look for answers and ideas 



COLLOIDS AND THE SOIL SOLUTION 7 

they can understand. Many current ideas about health, nutrition, and organic farming 
and gardening, for example, are popular not because they have been tested, but be- 
cause their simplicity is easily understandable and because not-always-sound logic 
makes them look good. Although science cannot provide all the answers, our lives 
are healthier and longer because we have broken away from deductive reasoning. 
The spurious logic, oversimplification, and rejection of careful testing that can be 
part of deductive reasoning are steps backward rather than forward. 



1.3 SOIL-ION INTERACTIONS 

Solutes, electrolytes, and nonelectrolytes in the soil solution are the immediate 
sources of the elements required by plants for growth. This supply can be continu- 
ously renewed by the many mechanisms of ion-soil interaction that remove and add 
ions in the soil solution: (1) mineral weathering, (2) organic matter decay, (3) rain, 
(4) irrigation waters containing salts, (5) fertilization, and (6) release of ions retained 
by the colloid or clay fraction of soils. 

Solutes in the soil solution and ions retained on soil particle surfaces are generally 
the largest fraction of the elements available to plants. Weathering of ions from soil 
particles is slow compared to plant needs. Organic decay releases ions much faster 
than weathering, but most of the ions released react with the soil's solid phase before 
they can be absorbed by plants or microorganisms. When retaining ions, soils strike 
a delicate balance between preventing losses by leaching and supplying plants and 
microorganisms. Ion retention by soils does not completely prevent leaching losses 
but is sufficiently strong that ions can recycle many times through soils, plants, and 
animals before they are finally lost to groundwaters, rivers, and the sea. 

Ions and molecules are retained in soils by cation and anion exchange, precipi- 
tation, weak electrostatic attraction, reactions with soil organic matter (SOM), and 
retention within microbial cells. If each ion were retained by only one such mech- 
anism and by only one soil component, soil chemistry would be relatively simple. 
Instead, each ion reacts by several mechanisms, and to varying degrees, with many 
different solid phases. For simplification, soil chemists generally either measure a 
single parameter that reflects most soil interactions with a given ion (e.g., its over- 
all availability to plants), or fractionate the soil and measure the ion's interactions 
with each soil component separately. Neither approach has proved totally satisfac- 
tory. Single-parameter measurements fail to account for variations from one soil to 
the next. Fractionation procedures may neglect the important interactions between 
soil components. 



1 .4 COLLOIDS AND THE SOIL SOLUTION 

The complexity of ion interactions with the soil's solid phase is greatly increased 
by the colloidal properties of the soil's clay and organic fractions. Colloids are sub- 
stances of about 1- to 1000-nm particle size that form unique mixtures when sus- 



8 INTRODUCTION 

pended in air or water. The components in colloidal mixtures tend to lose their indi- 
vidual identities so that the mixtures are like new substances. Colloidal mixtures are 
so ubiquitous and so distinctive that they have their own names: fog, smoke, smog, 
aerosol, foam, emulsion, gel, soil, and clay. All are small particles suspended in a 
liquid or gaseous fluid. Other colloidal mixtures include metal alloys, pearls, butter, 
and fine-grained rocks. The particles do not settle out of the suspending fluids, but 
also do not mix homogeneously with them. The physical properties of many colloidal 
mixtures tend to be similar regardless of the chemical composition of the colloid, be- 
cause the size and the interaction are so important. Colloidal suspensions of starch, 
soap, salad dressing, and clay in water, for example, look the same, are similarly 
viscous, similarly colored, opaque, and stable (do not separate into separate phases). 

Solutions are also mixtures but, in contrast to colloidal mixtures, solutions 
retain many properties of the major component — the solvent — while the minor 
component — the solute — loses its identity. Salts disappear when then dissolve in 
water. Aqueous solutions containing the same amounts of matter are as fluid and 
transparent as water. A mixture of 5% Na bentonite in water is a thick white gel; 
a solution of 5% NaCl is quite fluid and is transparent. When particles larger than 
ca. 2 fj.m are suspended in air or water, they settle out of suspension into separate 
phases. The properties of such mixtures are the sum of the properties of the separate 
components. 

Colloidal particles interact strongly with the fluid, but the individual particles have 
some structural integrity, so they cannot be said to dissolve homogeneously. The col- 
loidal mixture behaves so distinctively because of the large surface area of interaction 
between the particles and water or air. The ions at the boundary interact with the ions 
and molecules of both phases. This is true at any surface or phase boundary, but the 
interaction of colloidal phases is large because their surface areas are so large. A 
1 -mm sand particle has a surface area/mass ratio of about 0.002 m 2 g~ ' ; a 1 -£un clay 
particle, 2 m 2 g _1 ; and a 1-nm particle, 2000 m 2 g - '. 

The important colloidal properties that clays impart to soils include ion and 
molecular retention and exchange and water and gas adsorption. The colloidal prop- 
erties of clay create the intimate mixture of solids, liquids, and gases in the soil that 
is essential to life. 



1 .5 COMPOSITIONS OF SOILS AND PLANTS 

Except for carbon as CO2, H as HjO, and oxygen as O2, plants derive their essential 
elements from the soil. The minor exceptions are nitrogen and sulfur gases (NO A , 
NH3, and SO2) absorbed directly from the atmosphere by leaves, plus ions absorbed 
from dust and foliar sprays on the leaves. Foliar absorption can be significant in 
polluted atmospheres or where agricultural sprays are purposely applied. Under nat- 
ural conditions, the major factors affecting ion availability to plants are (1) the ion's 
concentration in the soil solution; (2) the degree of ion interaction with, and rate 
of release from, the soil's solid phases; (3) the activity of soil microorganisms; and 
(4) discrimination by the plant root during ion uptake. This book is concerned pri- 



COMPOSITIONS OF SOILS AND PLANTS 



9 



marily with the first two factors: the soil solution and ion interaction with the solid 
phase. 

Table 1.1 shows representative contents of important elements in soils. Soil con- 
tents vary and the values in Table 1.1 are averages. The composition of plants is less 
variable, partly because soil development tends to narrow the range of element avail- 
ability compared to the range of the elemental composition of rocks. The soil is an 
O-Si-Al-Fe matrix containing relatively small amounts of the essential elements. 
The matrix is virtually inert in terms of plant nutrition, but the small amounts of ions 
held by that matrix are vital. 



Table 1.1. Typical concentrations of essential elements in soils, ratios of plant ash to 
soil content, annual plant uptake, and the ratios of soil content (to 1 -meter depth) 
compared to annual plant uptake 











Soil Content/ 






Plant Ash 


Annual Plant 


Annual 




Soil Content 


Content/Soil 


Uptake 


Plant Uptake 


Element 


(Weight%) 


Content 


(kgha-^yr" 1 ) 


(yrs) 


Oxygen 


49 


— 


— 


— 


Hydrogen 


— 


— 


— 


— 


Silicon 


33 


0.3 


20 


21000 


Aluminum 


7 


0.03 


0.5 


180000 


Iron 


4 


0.1 


0.4 


100000 


Carbon 


1 


— 


■ — 


— 


Calcium 


1 


25 


50 


260 


Potassium 


1 


15 


30 


430 


Sodium 


0.7 


1 


2 


4600 


Magnesium 


0.6 


3 


4 


2000 


Titanium 


0.5 


0.08 


0.08 


62000 


Nitrogen 


0.1 


15 


30 


40 


Phosphorus 


0.08 


4 


7 


150 


Manganese 


0.08 


0.6 


0.4 


3000 


Sulfur 


0.05 


70 


2 


320 


Fluorine 


0.02 


1 


0.01 


26000 


Chlorine 


0.01 


10 


0.06 


200 


Zinc 


0.005 


5 


0.3 


2000 


Copper 


0.002 


5 


0.1 


1000 


Boron 


0.001 


50 


0.003 


400 


Tin 


0.001 


2 


0.001 


~ 10 000 


Iodine 


0.0005 


0.1 


0.00003 


22000 


Molybdenum 


0.0003 


~2 


0.003 


1000 


Cobalt 


0.0008 


1 


0.0006 


17 000 


Selenium 


0.000001 


~500 


0.003 


40 



From Vinogradov's data in N. F. Ermolenko. 1972. Trace Elements and Colloids in Soils. Israel Program 
for Scientific Translations, Jerusalem. 



10 INTRODUCTION 

Column 2 of Table 1 . 1 shows the ratio of plant content to soil content of important 
ions. The hydrogen, carbon, and oxygen ratios are omitted because these ions are not 
derived directly from soils. The ratios are crude indices of the relative availability of 
soil components to plants. Calcium, sulfur, nitrogen, and potassium in soils are more 
available than iron and manganese. One goal of soil chemistry is to explain why ions 
in soils vary widely in their degree of plant availability. 

Column 3 shows the approximate annual plant uptake per hectare (ha) of the el- 
ements, assuming an annual dry matter production of 10 000 kg ha - ' . The amounts 
of calcium, potassium, and nitrogen absorbed greatly exceed plant absorption of the 
other elements. 

Column 5 is the annual plant uptake of the elements divided by their total amounts 
in the soil. The result is the length of time the soil could supply that element to 
plants, if the plants were totally removed at harvest and nothing were added to the 
soil. The weakness of this assumption is that only a small fraction of plant matter 
is removed from soils during harvesting; most of the plant dies and decays where 
it grew. The assumption also ignores atmospheric and fertilizer inputs. Nonetheless, 
the ratios in column 5 roughly illustrate the relative size of the soil's store of essential 
elements. The soil's reserves of nitrogen, chlorine, and sulfur are low but are continu- 
ally replenished by microbial nitrogen fixation, gas absorption, and rain. Despite this 
replenishment, nitrogen and sulfur concentrations are often less than optimal for 
plant growth. Nitrogen is the element that most commonly limits crop productiv- 
ity. The low selenium ratio may be questionable because of limited data. The soil's 
supplies of potassium and calcium are also relatively low but are not cause for alarm. 
Calcium and potassium are constantly replenished by weathering of lower soil hori- 
zons and by decay of plant material, and they are easily replenished by liming or 
fertilization. The soil/piant conditions indicated in column 5 have existed for several 
thousand million years and are unlikely to change very much in the near future. 

The data in Table 1.1 are illustrative rather than quantitative. Soil has supplied the 
essential elements to living organisms since terrestrial life began. The Exhaustion 
Plot at Rothamsted Experiment Station in England, for example, has operated con- 
tinuously since 1845 and has shown that the worst agricultural practice — removing 
all plant material at harvest from the soil each year and no fertilization to make up 
the losses — reduces but does not stop plant growth or crop yields. 



1 .6 NONAGRICULTURAL SOIL CHEMISTRY 

Soil chemistry is also important to the nonagricultural uses of soil. Soil is a building 
material for earthfill dams and roads and is being rediscovered in industrial nations 
as a building material for homes. Brick and cement are, after all, only baked soil ma- 
terial. Lesser-developed nations have always recognized the livability of mud huts 
and adobe houses. The physical stability of soil structures depends in part on their 
soil chemical status. The longevity of mud and wattle construction in medieval Euro- 
pean homes and of the adobe buildings of Southwestern Native Americans depends 
on high calcium and low salt concentrations of the soils used to build those walls. 



BIOGEOCHEMICAL CYCLES AND POLLUTION 1 1 

The persistence through many centuries of the temples at Angkor Wat in Southeast 
Asia depends on the high concentrations of iron and aluminium oxides in the soils 
used to form the building blocks. Such laterite (now more properly called plinthite) 
materials from soil can dry irreversibly and thus resist slaking and weathering even 
in a humid tropical climate. 

The use of clay suspensions as drilling muds for lubrication and for clay liners for 
sealing landfills and lakes depends on the ability of Na + on colloidal surfaces to keep 
the clay from settling out and aggregating. The strong interaction of clay colloids and 
water is also useful in overcoming diarrhea, as a diluent for drugs, and as a drying 
agent. 

Many iron and aluminium ore deposits are the end result of long and extreme 
soil chemical weathering at the end of the plinthite stage. The time scale puts their 
formation into the category of geochemistry, but the mode of formation involves the 
same chemical reactions as soil formation. 

At the opposite end of the reaction time scale, soil clays are being investigated 
as catalysts to speed industrial chemical reactions. The "cracking" of petroleum into 
gasoline and other organic reactions is catalyzed by certain clays. Adsorption to clay 
surfaces imparts catalysis by holding reactive molecules in positions that encourage 
reaction and polymerization of organic chemicals. The Cu 2+ , Al 3+ , N(CH3>^ , and 
rhenium phosphine and rhenium phospho-organic complex ions adsorbed on clays 
have such catalytic properties. The large surface area of the clays produces such an 
intimate reactant-catalyst mixture as to be almost a homogeneous single phase. At 
the end of the reaction, the clay catalyst is easily removed by filtration. 



1 .7 BIOGEOCHEMICAL CYCLES AND POLLUTION 

The soil is a major part of the cycling of elements at the earth's surface. The elements 
that humans release as wastes are derived from the soil and the earth. Ions that are 
weathered at the earth's surface and released to the atmosphere or leached to the seas 
eventually circulate back to the land. They return to soils as gases absorbed from the 
atmosphere, as wastes removed from waste water, as solids buried in soils, as solutes 
in rain, as uplifted marine sediments, and as igneous rock uplifted to the land surface. 
Carbon, nitrogen, and sulfur cycle rapidly among the atmosphere, oceans, and soils. 
Other elements cycle more slowly between rocks, soils, and oceans but their move- 
ment is still rapid on a geologic time scale. The removal rate of elements from soils 
is slowed significantly by adsorption, precipitation, pH buffering, and plant uptake. 
Chemical pollution is the diversion of chemical elements from the natural biogeo- 
chemical cycles. The carbon, nitrogen, and phosphate in municipal wastes released 
to streams and lakes are removed from the soil-plant cycle, which is the source of 
the nitrogen and much of the phosphate. If those substances were instead put back 
directly into the soils from whence they came, much less pollution would result. Air 
and water only slowly convert their wastes back into their natural sites in plants and 
soils. Soil, on the other hand, has enormous surface area and microbial catalytic ac- 
tivity plus oxygen and water with which to deactivate pollutants. Soil degrades most 



12 INTRODUCTION 

wastes quickly and returns the components to their natural cycles, thereby minimiz- 
ing environmental disturbance. 

If one considers pollution to be the rendering of soils to be unfit for plant growth, 
then the greatest contribution to pollution is the salinization and urbanization of soils. 
Careless irrigation slowly adds increasing amounts of salts to soil, which can reduce 
and stop plant growth. Covering soils with asphalt and concrete also renders soil unfit 
for plant production. 

Human consumption of food, water, wood, metals, and fuel diverts substances 
from their natural cycles, if one assumes that humans are unnatural. Humans are 
trying to take care of themselves, just as other species do. The difference is that we 
concentrate wastes on a much larger scale and are better at using water and wind to 
carry them away. Humans, however, presumably have the capacity to analyze and 
improve their activities. This is hopeful and seems superior to ants, for example, 
which denude their environment and leave behind a totally barren land before moving 
to a new location. Even the worst of too-frequent shifting cultivation, monoculture 
cropping, clearcutting of forests, industrial pollution, and suburban sprawl are no 
worse than what ants do, and we have the ability to change and improve the situation. 

Consumption is the transformation of matter and energy into less useful forms, 
including dilution to concentrations less than those recoverable by our current tech- 
nology. Fertilizers, for example, are made from concentrated sources and diluted by 
spreading on agricultural lands. Over the short term, only a fraction is recovered by 
plants. The remainder is consumed by the soil when the fertilizer is converted into 
slowly recoverable or nonrecoverable forms. Not one atom is lost by this consump- 
tion, but the availability, chemical states, concentrations, and locations of the atoms 
change. 

Consumption of water includes discarding wastewater to the sea, where marine 
salts contaminate the water. It is usually much easier and cheaper to remove the 
small amounts of pollutants in wastewater than to remove the large amount of salt 
added to the water when it is dumped into the sea. Artificial desalination of seawa- 
ter is costly because of the energy required. Water is recovered naturally from the 
sea by solar evaporation. The resulting rain is unevenly distributed over the land so 
one-time use of water, in arid regions particularly, seems very wasteful. Water can 
often be used consecutively in the home, industry, and agriculture, thus decreasing 
water consumption. Wastewater spread on land can be agriculturally beneficial and 
simultaneously renovated for reuse. 



1.8 SOIL AND THE HYDROSPHERE 

The amount of water in soils is only a tiny fraction of the earth's total water supply. 
Table 1.2 shows two estimates of the distribution of water at the earth's surface. 
The two estimates differ widely but agree that the fraction present as soil moisture 
is small, 0.001 to 0.0005%. Either of these fractions of soil water seems perilously 
small to supply all terrestrial life and to help moderate climate. The periodic droughts 
around the world also emphasize that the amount of soil moisture is small. Assuming 



SOIL AND THE HYDROSPHERE 13 

Table 1.2. Two estimates of the distribution of the earth's water 

Ocean Water 80%" 97% A 

Ice 19 — 

Groundwater — 0.06 

Pore water in rocks 1 2 

Lakes and rivers 0.002 — 

Fresh — 0.007 

Saline — 0.007 

Soil moisture 0.001 0.0005 

(2 x 10 16 kg) (7 x 10 16 kg) 

Atmospheric water 0.0006 0.0001 

Biota — 0.0001 



21 br, i ■»<; x, i n21 



Total 1.7 x 10 ZI kg 1.35 x 10 2) kg 

" From R. M. Garrels, p. T. Mackenzie, and C. Hunt. 1 975. Chemical Cycles and the Global Environment. 
W. Kauggman, Los Altos, CA. 

''From D. A. Spcidcl and A. F. Agnew. 1982. The Natural Geochemistry of Our Environment. Westview, 
Boulder, CO. 



that the world's soils contain 10% water by volume, the 1.2 x 10 14 m 2 of terrestrial 
soils contain 1.2 x 10 13 m 3 of water to a meter depth. This imposing amount is the 
water that supports plants, weathers rocks, forms soil, and is the medium in which 
most soil chemical reactions occur. 

Terrestrial water receives most of its dissolved solutes from the soil, where rain 
first reaches the earth's surface and where weathering is strongest. The composition 
of water is less affected when it percolates to greater depths because the water already 
contains the salts obtained from the soil above. The composition and concentration 
of dissolved solutes can change at depths if the water contacts subsurface CaCO.-? 
or if it is stored for long periods in underground basins. In most cases, however, 
percolating waters retain the character of solute composition initially conferred on 
them by the surface soil. 

Stream water is soil drainage plus surface runoff. Natural drainage waters contain 
relatively low concentrations of the essential ions. This slow and steady input from 
drainage waters supports what is generally regarded as a natural and desirable aquatic 
population in streams and lakes. 

Runoff water is richer in sediments, nutrients, and organic matter. The fraction 
of surface runoff in temperate and humid regions is relatively small compared to 
drainage water under natural conditions. Dense plant cover prevents erosion and the 
soils are relatively permeable. Agriculture can increase the nitrate and phosphate 
concentrations, in runoff particularly, by removing the natural plant cover. Proper 
management can minimize runoff from agricultural lands, but some changes in water 
composition due to agriculture may be inevitable. Urbanization also increases runoff. 
The velocity of runoff increases as the overall permeability of the land surface is 
decreased, and runoff increased, by paving, compaction and destruction of natural 
contours. 



1 4 INTRODUCTION 

In arid regions, surface runoff is a considerable fraction of the stream flow. The 
intense storms, sparse plant cover, and relatively low soil permeability create inter- 
mittent streams "too thick to drink and too thin to plow." The invisible solute concen- 
tration in such waters is also high and can be as important to the downstream ecology 
as the sediment. 



1 .9 SOIL AND THE ATMOSPHERE 

The interaction of gases with soils is much less obvious than soil-water interaction, 
but is important to maintaining the low amounts of carbon, nitrogen and sulfur gases 
in the atmosphere. Gases that are foreign to the atmosphere are adsorbed by soils 
and plants and degraded to the natural and nongaseous forms of these elements. Soils 
also release gases, such as UzO and COi from organic decay and N2 and N2O from 
natural soil nitrogen compounds and from fertilizers. Soil has affected environmen- 
tal chemistry since the earth began, and the soil in turn has been affected by other 
components of the environment. 

Soil is a prominent part of the natural cycles of carbon, nitrogen, and sulfur. In the 
nitrogen cycle, nitrate and ammonium ions in rainwater are absorbed by soil, plant 
roots, and soil microorganisms and converted to amino acids or to gaseous N2 and 
N2O, which diffuse back to the atmosphere. Ammonia is also emitted and absorbed 
by soils. Under natural conditions the gaseous nitrogen loss is approximately bal- 
anced by N2 uptake and conversion to amino acids by symbiotic and free-living soil 
microorganisms. 

Soil absorbs sulfur dioxide, hydrogen sulfide, hydrocarbon, carbon monoxide, ni- 
trogen oxide, and ozone gases from the air. The reactions are subtle and are often 
forgotten in considerations of the composition of the atmosphere. Direct soil absorp- 
tion is perhaps most obvious in the case of the rapid disappearance of atmospheric 
sulfur dioxide in arid regions. The basicity of arid soils makes them an active sink for 
acidic compounds from the atmosphere. The relative amount of direct soil absorption 
of atmospheric gases, inappropriately termed dry fallout by atmospheric scientists, 
is less in humid regions where plant absorption and rain washout of the gases are 
substantial. 

The soil's role in the carbon cycle is very large. Table 1.2 shows the estimated 
amounts in the active carbon reservoirs at the earth's surface: the atmosphere, living 
biomass, freshwater, and ocean water above the thermocline (a temperature inversion 
that isolates the surface 50 m from the deeper ocean). The amount of soil carbon far 
exceeds the others. The emission of CO2 by organic decay in soils is the largest 
CO2 input to the atmosphere. The change in amount of organic soil carbon resulting 
from climate changes must have affected atmospheric CO2 concentration. Varying 
organic decay rates in soils due to climate and atmospheric CO2 changes have not 
been considered adequately in most discussions of the carbon cycle. A slower decay 
rate would have the same net effect on atmospheric CO2 that direct soil absorption of 
CO2 from the atmosphere would have. Peat (organic-rich soil) has been accumulat- 
ing in Canada and elsewhere since the glaciers retreated at an average rate of about 1 



SOILS AND THE DEVELOPMENT OF LIFE 1 5 

mm yr" 1 . Virgin soils, when cleared and cultivated, on the other hand, lose one-third 
to one-half of their original organic carbon content by oxidation to CO2. This amount 
of CO2 production over the last 100 years equals or exceeds the highly publicized 
amount released by fossil fuel combustion. 

Great changes in soil organic carbon levels have occurred during geologic time. 
The enormous carbon accumulation by soils during the Carboniferous Era must have 
affected the CO2 content of the atmosphere greatly. Carbonate accumulation and loss 
in soils is probably a smaller buffer of atmospheric CO2 than is organic soil carbon. 
The mass of soil carbon as carbonate is less, and its turnover rate is slower. 

The atmosphere, in turn, affects soil development by providing oxygen and by 
wind erosion and deposition. Sand dunes are only the most obvious example. Loess 
soils are deposits of silt-sized particles carried by winds from riverbeds and glacial 
outwash. A large fraction of the clay content of the soils along the eastern shores of 
the Mediterranean Sea has been carried by winds several thousand kilometers from 
the Sahara Desert and Atlas Mountains of North Africa. Trade winds carry Saharan 
clay particles several thousand kilometers out into the Atlantic Ocean. 



1 .1 SOILS AND THE DEVELOPMENT OF LIFE 

Soils have had a large role in the development of life and probably in the origin of 
life. Plants obviously have adapted to the physical and chemical characteristics of 
soils. These characteristics could also have aided the first steps of chemical organi- 
zation that preceded life. 

The earth's early atmosphere was rich in CH4, H2, CO2, CO, NH5, N2, and H2S, 
and contained no O2. In the late 1940s Miller and Urey showed that lightning dis- 
charges in such a gas mixture could produce small amino acid-like and other organic 
molecules that could be the organic precursors needed for life. The next question is 
how do these molecules polymerize to larger molecules? 

In the late 1920s a Soviet scientist noticed that hydrophobic substances in sea- 
water cluster together, like the tar balls familiar to beachgoers. He suggested that 
this could have been a step in the organization of simple organic molecules into the 
more complex ones of living organisms. One problem with this hypothesis is that 
the molecules formed in lightning discharge are hydrophilic. They dissolve in water 
and disperse in the oceans rather than accumulate as do hydrophobic hydrocarbon 
globules. 

In the 1930s the British biologist Haldane, inspired by this idea and rather igno- 
rant of soils, thought that the initial polymerization of small molecules might have 
happened in an organic-rich ocean, which he called the "primordial (pre-life) soup." 
This phrase caught the public's attention even though the source of the simple organic 
molecules was then unknown. To make even a thin 1 % soup in the approximately 
1.4 x 10 21 kg of ocean water requires about 1 x 10 19 kg of organic carbon com- 
pounds. The total amount of organic carbon on earth is estimated to be 8 x 10 18 kg, 
but these estimates were unavailable in Haldane's time. All of the carbon existing as 
methane and carbon dioxide in the atmosphere would have had to combine into small 



16 INTRODUCTION 

organic molecules to create this primordial soup. The oceans would disperse these 
organic molecules rather than concentrate and orient them into structural polymers. 

Bernal pointed out soon after that the adsorption by clays in soils and sediments 
would be important to concentrate and polymerize simple molecules. Although wa- 
ter neither concentrates nor catalyzes polymerization, his ideas were incorporated 
into the marine origin of life by marine workers' suggestions that the polymerization 
occurred in drying tidal pools. The drying also concentrates NaCl, which is inimi- 
cable to life. The marine biologists were eager to attribute as much importance to 
their discipline as possible. Evolution is opportunistic and would take advantage of 
the best chemical and physical conditions available for retaining, accumulating, and 
polymerizing organic compounds. These conditions are much more better realized 
in soils than in ocean water. 

Clay particles formed before life began, so primordial soils had essentially the 
same chemical and physical characteristics as today's soils. Clays could adsorb and 
concentrate simple organic molecules as they fell in rain, or adsorb them directly 
from the atmosphere. Amino acids, for example, have been shown to polymerize 
when adsorbed on chy surfaces; benzene and phenol polymerize spontaneously on 
Fe(III)- and Cu(II)-coated clays. Whether such reactions actually led to the origin 
of life is speculation, but these reactions are much more likely in soils than in tidal 
pools. 

In addition to concentrating organic molecules and catalyzing their polymeriza- 
tion, soils have several other properties that might be important in the origin and 
development of life: (1) protection of organic molecules from breakdown by ultra- 
violet light; (2) periodic drought, which encourages organic condensation (reactions 
that form water); (3) the relatively constant chemical composition of soil surfaces 
and the soil solution, much more so than that of the parent rocks; (4) higher avail- 
ability of phosphate and microelements than in water; (5) the much lower osmotic 
potential of soil solution than marine water; and (6) a cation-rich solution. Living 
organisms produce mostly organic anions, which require cations for balancing and a 
great amount and variety of cations than anions. These factors suggest that life could 
have begun more easily in the soil than in the oceans. 

Furthermore, the order of cation composition in plants and animals is Ca > K > 
Na = Mg and is close to the order of availability in soils, Ca > Mg > K = Na. 
The order in living organisms is quite different from the composition of seawater, 
Na > Mg > Ca = K. The amounts of solutes in body fluids of plants and animals 
is also closer to that of soil solutions than of seawater. The osmotic potential of 
body fluids is lower than that of seawater. Terrestrial plants and animals therefore 
die by ingesting seawater; marine plants and animals survive by excluding NaCl at a 
considerable expense of energy. As a result, life is more active in and on terrestrial 
soils than in the sea. Indeed, the open sea is barren. 

The concept of life originating in marine or tidal pools nonetheless persists and is 
supported by the rather complete fossil record in marine sediments. The fossil record 
on land is erratic, but this should be expected because of the rapid turnover of land 
surfaces compared to marine sediments. Soils have eroded, weathered, and dissolved 
into soluble salts into the sea many times since life began. A complete fossil record 



THE ROLE OF SOIL IN THE ENVIRONMENT AND THE MAINTENANCE OF LIFE 1 7 

under these circumstances would be highly unlikely. Much of the ocean bottom has 
remained intact, so fossils can accumulate readily and sequentially. 



1 .1 1 THE ROLE OF SOIL IN THE ENVIRONMENT AND THE 
MAINTENANCE OF LIFE 

The role of soil in the origins of life is controversial. Its role in maintaining life and 
the environment is clear, but often unrecognized and taken for granted. Soil scientists 
are partly responsible because they have been too modest and too reticent about soil's 
control of the atmosphere and hydrosphere. The atmosphere, biosphere, and hydro- 
sphere are weakly buffered against change and can fluctuate wildly when perturbed. 
Soils strongly resist chemical change and are a steadying influence on the other 
three environmental compartments. Changes in the hydrosphere, atmosphere, and 
biosphere due to human activities are often the result of bypassing the soil. The soil 
is the most robust environmental compartment. Treating solid, liquid, and gaseous 
wastes in soils before release to the hydrosphere or atmosphere can minimize envi- 
ronmental change. The soil is the source of most human wastes and ought to be site 
of their disposal. 

The environment is large and very complex, too complex for anyone yet to un- 
derstand fully. Scientific training necessarily tends to specialize — learning more and 
more about less and less. Scientists try to expand that myopic background to the 
whole environment, with mixed results. Among other things, they bring along bi- 
ases, of which one is usually that their background field is the most important. At- 
mospheric scientists, for example, naturally believe that the atmosphere is the most 
important part of the environment. The authors of this book are no different. We 
argue without apology that the soil plays the central and dominant role in the envi- 
ronment (Fig. I . I ). 

Table 1 .3 shows which of the four environmental compartments store significant 
amounts of the essential elements. Table 1 .4 shows the estimated quantities of the 
elements that circulate rapidly through the environment — C, fixed N, P, S, and water. 
The amounts shown for the oceans are those above the thermocline, a temperature 
inversion at about 50 m depth, which separates the surface water from the deeper 
water. The soil is the largest reservoir of almost all the essential elements and is the 
only reservoir for most of the essential elements. 

The general chemical and related physical properties of the four active environ- 
mental compartments are summarized as follows: 

Atmosphere. Poorly buffered against chemical changes; high amounts of readily 
available C as CO2; high CH; H2O variable; otherwise chemically uniform 
worldwide; entrained dust is the only reactive surface; high mass transfer rates. 

Biosphere. High C, N, and P concentrations; energy rich; high energy and mass 
transfer rates; high reactivity; ratio of biomass/soil is variable and climate de- 
pendent; complex and highly ordered structures. 



01 

o 

CO 

r 

3 
0) 
CO 



0) 
0) 



c 



■o 

c 

(9 



3 
(I) 

C 
0) 

p 



c 
o 
J] 

w 
CO 

i> 

o 

c 
o 



5 






ID 



BOM 

* 2 

O x 



00 

•2 2 
"3 o 

C/l — • 



O ~~ 

b o 
Z x 



e J? 

S3 o 
u x 







T3 


0> 






JS 


r0 




^-N 


O 


S 


,— . 4 


o 


.O 




o£ 


S3 
u 


'3 


^-^ 


•*_• 






8 


g 


© 
o 


s 


o 


© 


IO 


o 


8 


§ 




o 


o 


<5 






— 1 


© 







c c 

op go 

O o o 



a» 



Tt 


o 


<S 


o 
cm 




,__ 


^ 






_u 


U 


o 












'5 
a 


1 


c 


<N 


oo 


00 


po 


Z 


£ 


t-l 


s 


■w ' 






v — " 


o 


_ 


TT 


o 


o 






00 


o 






r— 1 


o 








o 








00 









So o o 
© o o 
t^ oo u-> oo 



•s 



o 



a) 

s 

t 

K I 
js o 

O a 
< CO 



a. 

c« I- 1 
IS _. 

S E 

O _!< 

S o 
00 " 

:s o 



18 



CHEMICAL UNITS 19 

Table 1.4. Estimated annual gains and losses from soils worldwide, "(rates before 
human influence, if significantly different, are in parentheses) 





Carbon 


Nitrogen 


Sulfur 


Process 


(Tgyr- 1 ) 


(Tgyr" 1 ) 


(Tgyr- 1 ) 


Soil organic matter decay 


-35 000 


— 


— 


Cultivation of virgin soils 


-1000 to -2000 


-100 to -200(0) 


-10 to -20(0) 


Peat accumulation 


200 


— 


— 


Soil nitrogen fixation 


— 


200 


— 


Atmospheric Nj fixation 


— 


4 


— 


NO* (mostly N2O), net 


— 


-8 to -25 


— 


Ammonia, net 


— 


-20 to -60 


— 


Nitrogen fertilization ( 1 970) 


— 


30(0) 


— 


Sulfur emission from flooded soils, net 


— 


— 


-3 


Soluble or eroded to the sea 


100 to 200 


-0 to -20 


-60 


Sea spray and solutes 


100 to 200 


20 to 20 


20 


Weathering 


— 


— 


40 



"Largely from B. H. Svcnsson and R. Soderlung, eds. 1976. Nitrogen, Phosphorus and Sulphur— Global 
Cycles. SCOPE Report 7, NFR, Stockholm. 

*l-or comparison, the carbon released by fossil fuel combustion in 1975 was 3000 Tg yr" . 

The sulfur released by fossil fuel combustion in 1 975 was 1 30 Tg yr~ ' . 

Hydrosphere. Poorly buffered against chemical changes; moderate O2 availabil- 
ity; low water availability because of high osmotic potential of seawater; low 
ratios of biomass/unit area and photosynthesis/unit area (because of little light 
below the surface); low temperatures; low essential elements (especially P and 
Fe); high NaCl. 

Soil. Chemically stable; strong buffering of elemental availability; variable H2O 
availability; soil solution is cation-rich; active microbial and inorganic cataly- 
sis; source and sink of all the elements essential to living organisms; high P but 
low availability; ratios of biomass/unit area and photosynthesis/unit area vary 
widely; photosynthesis is seasonal; high mass and energy transfer rates. 

1.12 CHEMICAL UNITS 

The number of atoms, not their mass or volume, determines the extent of a chemi- 
cal reaction. The unit of numbers is the mole, Avogadro's number (6.02 x 10 23 ) of 
molecules, ions, electrons, and so on. The atomic weight and molecular weight are 
the mass in grams (g) of \ mote of that substance. For example, one mole of H2 
has a mass of 2 g and contains 6.02 x 10 23 H2 molecules and 12.04 x 10 23 hydro- 
gen atoms. Ions have the same mass as their atoms, because the mass of electrons is 
insignificant. The numbers of moles of reactants and products can change during a 
chemical reaction, but the number of ions and the mass remain constant. 



20 INTRODUCTION 

The atomic weight of a chemical element is usually not a whole number, because 
it is the weighted average of the natural isotopes of the elements. Hydrogen, for ex- 
ample, has three isotopes: normal hydrogen (JH), deuterium (^H), and tritium (,H). 
The superscript is the atomic weight of the isotope, the number of protons plus neu- 
trons in the nucleus. The subscript is the atomic number, or number of protons, which 
chemically distinguishes each element. Deuterium and tritium are rare, so the distri- 
bution of isotopes in natural hydrogen yields an average atomic weight of 1.008 g. 
The atomic weights of the elements come closest to whole numbers when the natural 
mixture of carbon isotopes is assigned a mass of 12. The atomic weights and atomic 
numbers of me elements are shown in the periodic table (Table 1.5). 

Gravimetry, measuring the mass of products and reactants, is the most accurate 
form of chemical measurement. Aqueous solutions are more conveniently measured 
by volume, as molarity (M), the number of moles per liter of the solution. Sometimes 
molality, moles per 1000 g of solvent, is preferred for concentrated solutions, such 
as sea water or when temperature is far from 25° C. 

A 1 M (one molar) solution of CaCl 2 is 1 mole (147.03 g) of CaCl 2 • 2H 2 
dissolved in water and the mixture made up to 1 liter (L). The mixture is made 
up to volume after adding the solute, because salts and miscible liquids change the 
volume of water when added to it. Mixing 50 mL of ethanol with 50 mL of water, 
for example, yields about 80 mL of mixture. Salts also change the volume of water, 
but less dramatically. 

The numbers of moles of products and reactants may be unequal, as in 

H 2 S0 4 = 2H + + SO;* - (1.1) 

The numbers of moles of ion charge (formerly chemical equivalents), however, are 
equal in the equation. Equation 1.1 means, "one mole of sulfuric acid yields two 
moles of hydrogen charge and two moles of sulfate charge." Moles of ion charge are 
the number of moles multiplied by (1) the number of moles of H + or OH - that react 
with 1 mole of the substance, or (2) the number of "moles" or Faradays of electrons 
that 1 mole of the substance accepts or donates. The volumetric unit is molarity of 
ion charge, or moles of ion charge per liter. 

Low concentrations are sometimes expressed as negative logarithms, the "p" 
scale. This is most familiar as "pH," the negative logarithm of the H + ion concentra- 
tion: 

pH = -log(H+) (1.2) 

The p scale is also used for other ions — pNa, pH 2 P04, and pCa — and for the negative 
logarithm of equilibrium constants (pK) and solubility products (pK sp ). The negative 
logarithm increases as the concentration or value decreases. 

High concentrations of components in solid, liquid, or gaseous mixtures can be 
more conveniently expressed as mole fractions. The mole fraction is the ratio of 
moles of a substance to the total moles of all substances in the mixture. 

Gas concentrations are usually expressed as mole fractions or partial pressures. 
Partial pressure is a volume/volume ratio, and is equal to the mole fraction, at low 



O V 

■3 I 



u 

I 

xz 
u 



01 
■5 
3 

o 

I 

<0 

I 



E 




3 




C 




O 




£ 













TO 




£ 









s 









a 













a 













E 





















a 








n 








c 









E 




0) 




<u 




-C 


,**, 





(n 


41 








O 


a. 


TO 


c 


n 


CO 


■0 


cn 


a> 




r 













at 


c 






c 


(0 


a> 


Q> 


E 


CO 


0) 


O? 


0) 


c 


0) 


r> 


.c 

■s 


■a 


a> 


s 


XI 
(0 


c 



O 
'■5 


F 


O 










2. 


•0 

c 




CO 


m 








v- 





<l> 


n 


£1 


E 


<n 


>. 


r- 


in 



h- to 



0> o 

at 4* 



Z ^ ■*» 



£ < ^ 



£<S 



y m k> 



JL= 



- ^ - 



CM cr» 



9 ro 






K1 O-O 

01 S a> 



01 M ^ 






cm CL 



8 o.?. 



o 



> _ o 



25 



? 5 - 



CO 












?3 


IO 






»^- 








O 








K >■ 


r"-> 






O 
















Ol 








*> c 
















o> 








u> 








"> 








- UJ 


tO 






a> 








u> 








<ft 








5? 


K) 














u> 








<A 










rO 






S 








o> 








£^ 


K> 






- r- 








«r> 








ID 








C\J 




r- 




t-- 






6 ,, 


- O 




— 





1 




£ 




O 




rO 




Oj 




V 


!« 


- UJ 


ro 


0J 


ro 








<X> 




01 
















<\J 




- to 


rO 




Z3 

a » 






^ 




y? 




01 




















£ e 


lO 


CM. 


a. 

2 «n 














01 








CD 




<9 TJ 


f> 


rO 
CM 


3 <0 


z 
















UD 




ffi 












01 








O «_ 




fw 




5 Q. 


fl 


' w 


p in 

a. 


01 












CT» 




_ 




<M 












<j V 


V 




f * 


O 


m 




<0 












Oi 





00m 3d 



21 



22 



INTRODUCTION 



Table 1.6. Some SI units adopted by the Soil Science Society of America 











Former or 


Quantity 


Unit 


Symbol 


Definition 


English Units 


Length 


Meter 


m 




= 3.281 feet, 39.87 inches, 
10 6 microns, 10 10 
angstroms 


Area 


Square meter 


m 2 








Hektar 


ha 


10 4 m 2 


= 2.47 1 acres 


Volume 


Cubic meter 


m 3 








Liter 


L 


10~ 3 m 2 


= 1.057 U.S. quarts 


Mass 


Kilogram 


kg 




= 2.204 pounds 


Time 


Second 


s 






Energy 


Joule 


J 


kg m s -2 


= 0.2390 calories 


Force 


Newton 


N 


kg m s -2 


= 10 dynes 


Pressure 


Pascal 


Pa 


Nm -2 


= 10 -5 bars, 9.87 x 
I0~ 6 atm 


Power 


Walt 


W 


Js- 1 


= 1.341 x I0 -3 
horsepower 


Electric Current 


Ampere 


A 




= coulombs/second 


Potential difference 


Volt 


V 


JA-U- 1 




Resistance 


Ohm 


Q 


VA' 1 




Conductance 


Siemens 


S 


a 


= 1 mho (1 mmho/cm 
= ldSm" 1 ) 


Charge 


Coulumb 


C 


As 


= 1/(96516) faradays 


Amount of 


Mole 


mol 


6.03 x 10 23 


Avogadro's number 


substance 






entities 






Moles of ion 


mol^ or 


mol x ion 






charge 


mol ( -> 


charge 


= equivalent 


Temperature 


Kelvin 


K 








Celsius 


°C 




0°C = 273.13 K 


Radioactivity 


Becquerel 


Bq 


s-' 


= 2.7 x 10 _ " curies 


Absorbed dose 


Gray 


Gy 


J kg-' 


= lOOrad 


Concentration 


Moles per unit 




mol m 




Liquid 


volume 


M 


molL -1 
mol (+) L"' 
mot^L -1 


= molarity 

= normality of cations 

= normality of anions 




molecular weight 




kg" 3 


= 1000 ppm 




unknown 








Gas 


Moles per unit 

volume 
molecular weight 


mol m -3 




= 10 6 mg/m 3 




unknown 


kgirr 3 




= 10 6 mg/m 3 




mole fraction, or 


3 1 
m m- or 








volume per 


mol mol -1 




= 10 6 ppm(v/v) 




volume 









CHEMICAL UNITS 23 

pressures, where gases behave ideally. The volume of a gas is independent of other 
gases and is 22.4 L mol -1 at 25° C and 1 atmosphere pressure. The volume of each 
gas in a mixture is proportional to the number of molecules or moles in the gas. 
Low gas concentrations are often given as parts per million (ppmv), meaning the 
number of molecules of the gas per 1 million molecules of the gas mixture. Ppmv is 
a volume/volume and mole/mole unit for gases. (Ppm is a mass/mass unit for solids 
and liquids.) Ppmv in gases can be converted to mass/volume units by 

_■> molecular weight 

mg m = — — x ppmv (1.3) 

24.06 

The metric system was modified as Systeme International (SI) units (Table 1 .6) to 
prevent some confusion. The SI is based on seven fundamental units — including the 
mole, meter, kilogram, and second — from which the others are derived. The signif- 
icant changes for soil chemistry are mole of ion charge for equivalent, Siemens for 
mho, joule for calorie, and pascal for pressure. Table 1 .4 summarizes the SI units 
most frequently encountered in soil chemistry. SI allows easier conversion and com- 
munication between disciplines, but unfortunately discards some useful and familiar 
units, such as angstrom and equivalent. 

SI also recommends changes in the writing of numbers and units: avoiding the 
solidus (/), using the seven basic units in the denominator, using prefixes for large and 
small numbers (Table 1.7), and standardization of the decimal point. The solidus is 
removed by giving units in the denominator a negative exponent, so m/sec 2 becomes 
m sec -2 . Large numbers should be written without punctuation and with a space for 
each thousand — twenty five thousand is 25 000. Periods and commas in numbers are 



Table 1.7. SI prefixes 

Prefix Multiple Symbol 

Exa 10 18 E 

Peta 10 15 P 

Tera 10 12 T 

Giga 10 9 G 

Mega 10 6 M 

Kilo I0 3 k 

10 2 h 

10 1 da 

Itr 1 d 

1(T 2 c 

Milli 10 -3 m 



Hekto 
Deka 
Deci 
Centi 



Avoid 

when 

possible 



Micro 10 -6 \i 

Nano 10~ 9 n 

Pico 10 _ 12 p 

Femto I0 -15 f 

Alio 1CT 18 a 



24 INTRODUCTION 

confusing: the English and American 2,500 means l\ to Europeans. In SI the period 
and comma both denote the decimal point; 2.5 and 2,500 mean 2 ^ in SI. 

For soils in the field, soil area is usually more convenient and more meaningful 
than soil mass, but the square meter is too small and the square kilometer too large. 
The hektar or hectare (ha = 10 4 m 2 ) has been adopted as an area unit. To convert soil 
mass to area, the common factor for mineral soils is 2 million kg ha -1 15 cm -1 of 
soil, assuming a soil bulk density of 1300 kg m -3 . Cultivated organic soils have bulk 
densities of about 300 kg m~ 3 . 

ADDITIONAL READING 

Bolt, G. H. (ed.). 1990. Interaction at the Soil Colloid-Soil Solution Interface. Kluwer, Dor- 
drecht. 

Cresser, M. S., K. Killham, and T. Edwards. 1993. Soil Chemistry and Its Applications. Cam- 
bridge University Press, Cambridge, UK. 

Huang, P. M. (ed.). 1998. Future Prospects for Soil Chemistry. SSSA Spec. Pub. 55, American 
Society of Agronomy, Madison, WI. 

McBride, M. B. 1994. Environmental Chemistry of Soils. Oxford University Press, New York. 

Russell, E.W. 1987. Soil Conditions and Plant Growth, 1 1th ed. A. Wild, ed. Longman, Lon- 
don. 

Sposito, G. 1994. Chemical Equilibria and Kinetics in Soils. Oxford University Press, New 
York. 

Tan, K. H. 1998. Principles of Soil Chemistry, 2nd ed. Dekker, New York. 

QUESTIONS AND PROBLEMS 

1. Starting with a cube 1 m on a side, calculate the change in surface area by sub- 
dividing it successively into cubes 1 mm, 20 fj,m, or 1 /itm on a side. How many 
particles would be in each size group? 

2. What is the mass of each of the above size particles'? The specific gravity of alu- 
minosilicates is 2.65. Calculate the surface area/mass ratios of the above cubes, 
assuming that they are aluminosilicates. 

3. What were the shortcomings of van Helmont's experiment and how could it have 
been improved? 

4. Compare the lists of macroelements for plants and for animals. Do the same for 
the microelements. Which are derived directly from the soil? What is the soil's 
role, if any, in providing the essential elements not directly supplied by the soil? 
Which elements are influenced significantly by the atmosphere and by humans? 

5. Which of the essential elements are metals? Nonmetals? Cations? Anions? 

6. Why was it necessary that the macroelements be of low atomic weight for life 
to evolve as we know it? 



QUESTIONS AND PROBLEMS 25 

7. Early in the earth's lifetime, the atmosphere was apparently rich in methane and 
hydrogen and low in oxygen. Lightning can create simple amino acids in such 
a gas mixture. Describe the sequence of reactions and the soil properties that 
might aid polymerization of these amino acids in soils. 

8. Discuss the differences between an element's extractability, "availability," and 
total content in soils. 

9. Discuss why CO2 exchanges between soils and the atmosphere. 

10. Calculate the amount of water held in the world's soils if the average soil mois- 
ture content is 20% by mass {the land area is 1.2 x 10 8 km 2 , the soil's specific 
gravity is 1.3, and the soil is 1 m deep). Compare this value with the soil water 
data in Table 1 .2. 



2 



IMPORTANT IONS 



This chapter discusses the individual chemical elements in soils. Most questions 
about soil chemistry are about specific ions and many people are satisfied with a 
quick, albeit incomplete, answer. The processes that govern these ions in soils and 
can yield a more complete answer are discussed in the later chapters. If the chem- 
ical symbols and nomenclature used for conciseness are unfamiliar, please refer to 
Chapter 3, where they are explained. 

As the earth formed and cooled during its origin, the lighter chemical elements 
tended to float to the surface. The earth's center is thought to be iron-rich and has a 
density of >6000 kg m~ 3 . The density of the outer crust, or mantle, is about 2800 kg 
m -3 . The density of the rock minerals at the earth's surface is about 2650 kg m -3 . 
The elements in the rock minerals at the earth's surface (Table 1.1), are the starting 
materials for soils and also contain the essential elements from which soil and life 
evolved. 

Soils contain all of the natural chemical elements in the periodic table, including 
the elements that are essential to living organisms and those labeled as toxic. Only a 
few elements make up >95% of the mass and volume of the earth's crust (Table 1.1). 
The remainder are present in soils in small but important amounts. Even the rarest 
natural element, the alkali metal francium, 87 Fr, which has never been isolated, oc- 
casionally appears as faint lines in the arc spectrum of arid soils where alkali metal 
sails accumulate. 

Only in a few areas, at least as far as we now know, are any elements naturally in 
a state that can cause harm. In a few other unfortunate areas, soils contain synthetic 
chemicals, synthetic radioactive isotopes, and elements that have been concentrated 
by human activities. These areas are a serious problem. Tn an agricultural sense, the 
severest concentration problem is the widespread accumulation of salts in irrigated 
arid lands. The mere presence of a chemical in soils is rather insignificant. What 

26 



IMPORTANT IONS 27 

matters is the availability of a substance to plants and to the soil solution; its total 
concentration is of much less significance. Many soils labeled as contaminated be- 
cause of their total element content are benign, or can easily be made benign, because 
their availability to plants and movement to groundwater is minimal. The working 
definition of a contaminated substance is that it will make you sick if you eat it, a 
poor definition to apply to soils. Soil processes tend to change the availability of 
elements added to soils back to that in native soils. The plants and groundwater on 
copper, lead, and zinc ore deposits are not necessarily contaminated, but the total 
amounts of those ions in the soils are very high. 

For most purposes, the important ions in soil chemistry are those that are essen- 
tial, or toxic, to living organisms. Although opportunistic, evolution took little or 
no advantage of the most prevalent ions — O, Si, Al, Ti, Na, and CI — in soils and 
oceans. Figure 2. 1 shows the essential ions in bold type, and the toxic ions are cross- 
hatched. The soil is both the source of the essential elements and the safest disposal 
site to return the elements back to their native biogeochemical cycles. The elements 
are present as ions in nature because the zero oxidation state is unstable for most 
elements. The exceptions are O2, N2, the inert gases, and the precious metals Au, 
Pi, Ag, and so on. In the elemental state atoms bind only to each other. As ions, the 
elements are active and react with other ions. 

In recent decades soil chemistry has increasingly been involved with pollution — 
unnaturally high concentrations of chemicals in the environment caused by industrial 
activities and by high population density, which concentrates wastes. Much of that 
harm could be mitigated by reacting those substances with soils rather than releas- 
ing the chemicals to the atmosphere or hydrosphere. The soil reactions that control 
natural concentrations in soils can also buffer the effects of large additions to soils. 

The distinction between essential elements and toxic elements is not clearcut. The 
essential ions have a concentration range in which they are essential for organisms 
to complete their life cycle. The boundaries of this range vary with the species of 
the organism. At concentrations below that boundary, organisms suffer deficiency. 
Above that boundary, increasing amounts do not increase growth, and if the concen- 
tration is well above the beneficial range, the essential elements can be toxic. "Toxic" 
elements are those that are more likely to be found at this excessive concentration, 
or elements that are toxic at low concentrations. The excessive concentrations are 
almost always due to human activity. Evolution has accepted the natural concentra- 
tions and availabilities of these elements in soils, water, and air. The availability is 
governed largely by their reactions in soils and sediments with the soil solution and 
fresh and marine waters. 

Some elements have incorrectly been labeled as all good or all bad. Oxygen at 
20 volume percent O2 in air is necessary for aerobic organisms but is deadly to 
anaerobes. Concentrated, 1 00%, O2 is helpful for humans to breathe only for a short 
time; longer periods are harmful. Ar.senic is known as a toxic element, but low As 
concentrations are essential to living organisms. 

All substances are toxic if their concentrations are too high. The toxic ions in 
Fig. 2.1 are distinctive because, as far as is yet known, they do not benefit liv- 
ing organisms at any concentration, and they are toxic at low concentrations in the 



a. 



■;:M 


"• : s; : 


■■;< 




X 


c 
or 


O 


















S 






a 


IL 


G 




H* 


5 


c 
o 

z 


3 
-1 




§ 


o 


V) 


« 


£ 


£ 


o 

2 


> 




< 


2 


Q. 


< 


to 


CO 




1 






u 


i/S 


3 


A 


1 


UJ 




i 


CO 


II 


3 


C 


t- 


* 




2 m 




11 


li 


a 




S H 




W^ 


^1 








CD 
n 


3 

o 


* 


3 
< 


£1 

1- ' 




- 


11 


S 5 


CL 


13 






N\\\\^ 












CO 

a 


S 




i-« 


3 
UJ 






£ 


3 

q: 


8 








CD 


c 






E ' 




a 


5 


i- 


£ 




0. 




00 


CJ 


i 


* 


z 


D"'-;. : : 


a 


> 


n 
Z 


£ 


CL 


' 3- '.'■'. 

Q- 


m 


V 


w 




0) 




H 


.: 1- 


N 


X 




o 


"h-v-.; 












10 


r- 




B 


•: « 


>- 


3 


<f 


s 

c 

o 


4> 


< 


11 


? 


S 


1/i 


£ 


s 


u 














o 




















i 


2 


•* 






u. 






- 


CJ 


IO 


■* 


m 


CO 


N 





<5 
u> 

c 
a> 

£ 



o 

e 

E 
o 



2 

(0 



E 


0) 

■= «j 
o> <a 



E>= 
a> o> 



•5| 

to <o 

£ 3 



s§ 

CO .c 

C C 

a> ~ 

© c 
£ a> 
~ u> 

jQ a. 

E g 

a; E 

a> £ 

a) c 

£ 43 

*o a> 

© E 

ro a) 






u. .£ 



IMPORTANT IONS 29 

food/water/air web. The list is biased because it refers mainly to humans, but other 
organisms probably have the same response as humans. Indeed, we assume this when 
we use laboratory animals for testing toxicity. 

With the important exception of plant toxicity by aluminium in acid soils, toxicity 
in soils appears to be due to human-caused conditions. The number of toxic ions 
would be much greater except that humans have not been exposed to other elements 
enough for health and regulatory concern. 

The "toxic" ions are present in all soils, plants and water, but the natural con- 
centrations and availabilities are so low that they appear to pose no danger to or- 
ganisms. Animals have an effective buffer against high concentrations because they 
ingest plants rather than soils. Plants are more tolerant of toxicity, absorb these ions 
in small amounts, and tend to retain them in the roots rather than translocate them 
to plant tops. This plus soil unavailability keeps toxic ions out of the animal food 
web. Isolating and assessing the health effects of the background concentrations of 
the toxic ions is difficult. Lead, for example, is ubiquitous in the environment but is 
retained strongly by soils, so the Pb concentrations in food and water are low. Some 
people worry about Pb in drinking water, but the amounts added to the human body 
by poor plumbing, flaking old paint, some folk remedies, and some pottery glazes 
are far greater. 

Although most of the elements in Fig. 2. 1 are shown as neither essential nor toxic, 
our knowledge is incomplete. The number of essential elements is probably, and the 
number of toxic elements is certainly, larger than shown in Fig. 2.1 Water, plants, 
and dust contain many ions, so the exposure of organisms to most of the elements in 
the periodic table is continuous, but in very small amounts. Proving their essentiality 
or toxicity under these conditions is difficult. To show the essentiality of chlorine to 
plants, for example, the plants were grown in greenhouses with carefully filtered air 
to remove dust and water droplets, the plants were grown hydroponically because 
even acid-washed quartz sand contains sufficient CI to satisfy the plant's small CI 
requirement, the water was doubly distilled and could not contact glass to avoid 
leaching CI out of the glass, and the plants had to be second generation grown in this 
environment. Normal seeds contained enough CI to satisfy this annual plant's life 
cycle requirement of CI. 

This experiment may have been of more than just academic interest. Because 
the CI input to most regions is marine CI entrained in rain, and because soils retain 
CI weakly, continental regions far from the oceans could have low CI in the soil- 
plant-water system. Chlorine deficiencies may occur, just as has been shown for Cl's 
chemical relative, iodine. The human disease goiter, due to I deficiency, was endemic 
in the interiors of the continents until the 20th century when I supplements in table 
salt and wider distribution of food fish became common. Low CI concentrations in 
soils may subtly affect plant and animal health. NaCl improves plant growth in some 
parts of Australia, although NaCl is best known as being harmful to plants at high 
concentrations. The Australian plants may use the Na to supplement a low K supply 
rather than to overcome a CI deficiency, but a suggestion of CI deficiency remains. 
The strong desire of animals and humans for more NaCl in their diet may be due to 
a subtle Na and CI deficiency as well as due to taste. 



30 IMPORTANT IONS 

2.1 ESSENTIAL ELEMENTS 

All organisms need the essential macroelements or macronutrients — C, H, O, P, K, 
N, S, Ca, Fe, and Mg — in relatively large amounts to complete their life cycle. A 
mnemonic phrase to remember them is "See Hopkins cafe, mighty good." Iron is 
needed in smaller amounts than the others, but is necessary for the mnemonic. An- 
imals additionally require Na and CI as macroelements, but since animal biomass 
is only 1/10000 that of plant biomass, animal requirements are insignificant in the 
overall picture. Macroelements have been called macronutrients for a long time, but 
"nutrient" implies energy content, as in carbohydrates. 

Organisms need the essential microelements or micmnutrients — B, Si, F, CI, V, 
Mn, Fe, Cr, Co, Ni, Cu, Zn, Mo, Ar, Se, Sn, and I — in smaller amounts than the 
macroelements, but they are just as essential. Whether all the microelements are re- 
quired by all organisms is unclear. Several have been shown to be essential for only 
one species. Because proving essentiality is so tedious and expensive, if an element 
is essential for one or a few species, it is assumed to be essential for other organ- 
isms as well. The list of essential microelements will probably grow as experimental 
techniques become more and more refined. 

The soil is involved with all the essential elements. Most are present as ions in 
the soil solution and flow into the plant as it absorbs water. Plants obtain hydrogen, 
carbon, and oxygen from water and air, but soils provide water-holding capacity, 
provide pore space for O2 and CO2 movement between plant roots and the atmo- 
sphere, and supply CO2 to the atmosphere through the decay of organic matter by 
soil microorganisms. 

Animals derive their essential elements from plants. The ability of plants to supply 
these elements to plants and animals depends on a combination of factors: availabil- 
ity of the ions in the soil solution, plant selectivity at the soil solution/root interface, 
and ion translocation from root to plant top. The system is good but not perfect. The 
concentrations of essential elements are occasionally too high or too low for ani- 
mals, because plants can tolerate a much wider range of elemental concentrations 
than can animals. Plant contents of iron, for example, tend to be lower than ideal for 
the human diet. Grazing animals in a few semiarid parts of North America may suf- 
fer from high selenium, in Australia from low cobalt, and formerly in Norway from 
low phosphorus availability because of unsatisfactory amounts in soils. The supply 
of essential elements by plants to animals is generally adequate; cases of too little or 
too much are noteworthy because of their rarity. 



2.2 TOXICITY AND DEFICIENCY 

The boundaries between toxicity, sufficiency, and deficiency are vague. The amounts 
vary with the species of plant and animal, vary within the species's growth cycle, 
and vary with the organism's general health and the supply of the other essential 
elements. The elements in the crosshatched areas of Fig. 2.1 are the toxic elements 
of primary concern to government regulatory agencies — Be, Cd, Hg, Pb, and all 



TOXICITY AND DEFICIENCY 31 

radioactive elements. As far as is presently known, these elements are harmful to 
humans and probably other organisms when present above typical natural concentra- 
tions, and are of doubtful benefit at any concentration. The nonradioactive elements 
are present in soils, but at low concentrations that have not yet been shown to affect 
life. Probably most of the unmarked elements in Fig. 2. 1 are "toxic," but exposure at 
above their natural concentrations is too rare to be a general health concern. 

Other elements that can be in possibly toxic concentrations in soils, water, and 
plants are Ni, Cu, Zn, and As. Toxic elements can be trendy and political as one 
or more reach popular attention. Regulations generally use total soil concentration, 
rather than soluble or plant-available concentration, as the criterion of toxicity be- 
cause defining "soluble" and "available" is difficult. The ash from nerve gas inciner- 
ation, for example, may have been incorrectly classified as a hazardous waste because 
it contained 50 mg kg -1 lead. The native soils in the area contained 70 mg Pb kg -1 . 
The stigma of the ash's origin as nerve gas waste probably had much to do with its 
being labeled a hazardous waste. Copper mine tailings contain 0. 1 % Cu — does that 
make them a toxic waste? The native ore contains 0.5 to 1% Cu and does not contam- 
inate groundwater. Opponents to fluoridation of water would add F to the toxic list 
because the water additive NaF is a rat poison at high concentrations. Plutonium is 
very toxic and was feared to be at significant concentrations in tobacco. The possibil- 
ity of the radioactive gas radon diffusing from soils into poorly ventilated basements 
recently raised much concern. Some groups are labeling CI as a dangerous element 
because of its effects on stratospheric ozone and want to ban CI2 as an industrial 
chemical. Since elements can be both beneficial and harmful, banning elements to 
create a completely risk-free environment is as sensible as banning fire. 

The other issue is told in the adage, "The poison is in the dosage." The amount of 
the substance is as important as the substance. NaF is a rat poison at high concentra- 
tions but it prevents tooth decay, without known side effects, at low concentrations. 
Under natural conditions toxic concentrations of the regulated elements Be, Cd, Hg, 
and Pb are very rare, but they are present in all soils and plants at low concentrations. 
They are the most likely elements, however, to be found at toxic concentrations un- 
der anthropogenic (human-caused) conditions, such as metal industries, mine spoils, 
and landfills. Overall, the dangers created by these elements are small compared to 
organic compounds — pesticides, drugs, allergens, and so on — but in local areas they 
can be serious. 

Toxicities of Be, Cd, Hg, Pb, and others have been reported where people and 
plants have been exposed directly, without the protection of the soil-plant system. 
Beryllium dust in industrial air, F from aluminium smelting and phosphate fertilizer 
manufacture, Pb and Cd from water pipes and from copper and zinc smelting, Se 
from drainage of high-Se soils, and a recent scare about Hg amalgams in tooth fill- 
ings are examples. Although public awareness of chemical pollution is increasing, 
exposure to toxic elements and substances is decreasing. People are living longer 
and healthier than ever before. The public media and alarmists to the contrary, part 
of that is due to decreasing exposure to toxic concentrations of chemicals. 

In the "good old days," lead water pipes were common and lead was in pewter 
tableware and in ceramic glazes. Lead arsenate was a pesticide spray on apples and 



32 IMPORTANT IONS 

other crops. Workers in Cu, Pb, and Zn smelters were exposed to extremely high 
amounts of those and other metals in those ores. People ate antimony salts in the 19th 
century to make their hair black and glossy. The Mad Hatter in Alice in Wonderland 
may not have been unusual then because Hg compounds were used to make felt and 
a symptom of Hg poisoning is unbalanced behavior. If these elements are in soils, 
they are generally retained strongly and kept unavailable to plants. If absorbed by 
plants, they tend to accumulate in roots and little is translocated to plant tops. Our 
gastrointestinal tract also insulates us to some extent from these elements in food and 
water. 

All substances are toxic above an ill-defined threshold concentration that varies 
with each substance and with each species. Under natural conditions and at our cur- 
rent state of knowledge, toxicity is rare. One exception is Se accumulation by legume 
plants in a few semiarid regions of North America. Se is toxic to grazing animals, but 
the legumes show no symptoms. Another example is occasional Mn and Fe toxicity 
to rice plants in acidic Asian rice paddies. The most prevalent natural soil toxicity 
problem is largely hidden from us. Aluminium toxicity to plants in acid soils is so 
widespread that we accept it, perhaps appreciate it, in forested areas. Forest plants 
are more tolerant of Al toxicity than food crops and grasses, which might other- 
wise compete with the trees in those areas. Because trees are beautiful and valuable, 
we accept the effects of Al toxicity until we want that land to grow food crops and 
grasses. 

Plants must grow in the environment within reach of their roots and leaves. To 
survive, plants have had to evolve a broader tolerance to imbalances of essential and 
toxic elements than have animals. Deficiencies and toxicities in plants are usually 
evident by reduced yield or reproductivity, abnormal coloration, and plant and fruit 
deformities. Animals can supply their more stringent nutritional requirements by 
ingesting food grown on a variety of soils. 

Deficiencies of soils to supply the essential elements are more obvious in plants 
than in animals. The nitrogen content of most soils is low enough that plants benefit 
from added N. In many soils, plants, especially food crops, also grow better with 
added P, K, and S. Because the response differs among plant species, fertilization 
can change the distribution of plant species. Since we usually define the untouched 
plant community as the ideal state, species changes are considered detrimental. 

Soil microbes lessen the changes due to fertilization because they compete more 
effectively than plants for ion uptake, but microbial changes are rarely noticed. 
Changes due to fertilization are more obvious in air and water, which cannot resist 
change as well as soils, or at least the changes are more obvious. Green and turbid 
waters teem with actively metabolizing algae and bacteria, but not with the larger fish 
that we appreciate. Hence, we call these waters polluted. Microorganisms degrading 
the dead algae and bacteria compete too strongly for oxygen in those waters for fish 
to survive. Only when the essential element and nutrient content in a water body 
is relatively low can fish compete effectively. When Lake Erie was "dead" years 
ago, due to high inputs of industrial and agricultural wastes, the total metabolic 
activity was much greater than it is now. The fish population, however, was low. The 
microbes consumed the wastes and, as the input decreased because industrial and 



TOXICITY AND DEFICIENCY 33 

municipal wastewater was cleaned, the microbes consumed themselves. Now fish 
again survive; the clear waters that we enjoy are the aquatic equivalent of a rather 
barren soil. 

The ability of soils and soil microbes to neutralize excess fertilizers has limits. Ni- 
trogen and phosphate draining from excessive fertilization of sugar cane fields favors 
cattail plants over native plants in the Florida Everglades. Nitrogen and phosphate 
from fertilizers and municipal wastewater are causing water problems worldwide. 

Iron anemia (Fe deficiency) in humans is common, but is due to low availabil- 
ity to plants rather than low amounts in the soil. Molybdenum deficiency, which 
prevents microbial nitrogen fixation, and cobalt deficiency in Australian sheep that 
prevents rumen bacteria from synthesizing vitamin B12, have been reported in Aus- 
tralia. Phosphate deficiency that led to weak bones in grazing animals was reported 
in Norway. 

Table 1.1 lists the total amounts of the essential ions in soils, including ions within 
the crystal lattices of clay, silt, sand, and gravel particles, which are unavailable by 
plants. Total concentrations are poor indicators of plant availability. A more useful 
value would be the amount of each ion available during the plant's growing season. 
Measuring this availability has been, and still is, the subject of much research. As- 
sessing availability is difficult because ion uptake by plants varies with plant variety 
and growth conditions as well as with soil chemistry. The major factor influencing 
an ion's availability is its soil chemistry, but differences in plant response are also 
important. 

One aid to developing a perspective of soil chemistry is to arrange the elements 
according to their behavior in soils. The elements of primary interest exhibit a wide 
range of chemical and biochemical behavior. Their essentiality or toxicity to organ- 
isms is evidence that the chemistry of each is unique. Their soil chemistry, however, 
is less distinctive, and their active fractions can be arranged into behavioral groups. 
Table 2.1a shows the important chemical elements organized into six behavioral 
groups: exchangeable cations, water-soluble anions, weakly water-soluble anions, 
weakly water-soluble transition metals plus aluminium, toxic ions, and redox-active 
elements. 

The major exchangeable cations neutralize and are retained by the negative charge 
of soil minerals and organic matter. The major anions are at lower concentrations in 
the soil solution than are the cations in most soils. These anions neutralize cations 
and the negative charge of soil particles. Weakly water-soluble anions are present 
at concentrations of < 10 -5 M in most soil solutions. The concentrations of the 
transition metal and aluminium ions in (he soil .solutions of typical agricultural soils 
are also in the < 10 -5 M range. This group includes essentially all of metals in 
the periodic table except for the alkali and alkaline earth metals. Toxic ions are of 
current concern in the literature. Aluminium is included in this group because of its 
widespread plant toxicity in acid soils. Many more transition metals would be in this 
group except that soil reactions markedly depress their availability to plant uptake 
and movement to groundwater. 

The last group in Table 2.1a contains the elements active in oxidation-reduction 
reactions in soils. Changes in oxidation state greatly change the chemical properties 



34 



IMPORTANT IONS 



Table 2.1a. Ions of major interest in soil chemistry, grouped according to major 
behavioral modes and shown as their most common states in soil solutions 



Ion 



Ca 2 + 

Mg 2+ 

Na+ 

K+ 

NH+ 

Al 3+ (H+) 



no: 



so 2 - 

cr 

hco; 



CO, 



Comments 



Major Exchangeable Cations 
Occur predominantly as exchangeable cations in soils; these 
ions are relatively easily manipulated by liming, irrigation, or 
acidification; exchangeable Al 3+ is characteristic of, though 
rarely the predominant exchangeable cation in, acid soils; pro- 
ductive agriculura! soils are rich in exchangeable Ca 2+ 



H 2 PO-, HPO^" 
H 2 As0 4 , As0 2 
H3BO3, H 2 BO~ 
Si(OH) 4 
M0O 2 " 

Al 3 +, A10H 2+ , Al(OH)+ 
TiOOH+ (?) 
Fe(OH)J, Fe 2+ 
Mn 2+ 



Major Anions 

Present in considerably lower concentrations than the major 
cations in all but the most coarse-textured and strongly saline 
soils, where they are essentially equal; sulfate and NO^" are 
important nutrient sources for plants; sulfate, Cl~, and HCO^" 
salts accumulate in saline soils; carbonate ions are present in 
appreciable amounts only in soils of pH > 9 

Weakly Soluble Anions 

Strongly retained by soils; borates are the most soluble of the 
group; retention or fixation by soils is pH dependent; molyb- 
date and silica are more soluble at high pH; phosphate is more 
soluble at neutral or slightly acid pH 

Transition Metals and Aluminum 

Insoluble hydroxides tending to accumulate in soils as silica and 
other ions weather; iron and manganese are more soluble in wa- 
terlogged or reduced soils 



(continued) 



of the elements. Table 2.1a shows the elements in their most common oxidation 
states in soils, including the likely ion species in soil solutions. The higher oxidation 
states are typical of the soil solutions and secondary minerals of aerobic soils. The 
lower oxidation states tend to be found in organic matter, flooded soils, and igneous 
minerals. 

Some elements appear in more than one group because they have more than one 
important function. In addition, chemical properties overlap. The arrangement in Ta- 
ble 2.1a implies only that the elements in each group exhibit similar, rather than 
identical, behavior in soils. Table 2.1a does little to indicate the relative importance 
of the chemical elements. Certain elements dominate soil reactions because of their 
greater abundance, because of the rapidity of their reactions, or because they are a 
source of energy. For example, all transition metal ions and aluminium produce acid- 



ALKALI AND ALKALINE EARTH CATIONS 



35 



Table 2.1b. (Continued) 



Ion 



Comments 



Cu 
Zn 



2+ 
2+ 



Cd 2+ , A\ 3 + 



Pb 



, 2 + 
,2+ 



Hg z+ , Hg 



Be 



,2+ 



AsOr, CrOl 



C (organic to HCO^ ) 
O (O 2- to 2 ) 
N(— NH 2 toNO^) 
S(— SHtoSO 2- ) 
Fe (Fe 2+ to FeOOH) 
Mn(Mn 2+ toMnOi. 7 ) 
Se (organic to SeCr 2- ) 
Hg (organic to Hg° or Hg 2+ 



More soluble than the above cations in all but very acidic soils; 
availability increases with increasing soil acidity; complexed 
strongly by SOM 

Toxic Ions 

Soil behavior similar to transition metals; Al 3+ is a hazard to 
plants; the others are of more concern to animals; CD 2 " 1 " is rel- 
atively soluble, available to plants, and its retention is relatively 
independent of pH; remaining ions are less available to plants 
with increasing pH, except perhaps for As; the last three ions 
have received relatively little study in soils 

Active in Oxidation-Reduction Reactions 
Soil biochemistry revolves around the oxidation state changes 
of soil carbon, nitrogen, and sulfur compounds; molecular oxy- 
gen is the main electron acceptor; FE(UI), Mn(lII-lV), nitrate, 
and sulfate are electron acceptors when the oxygen supply is 
low 



ity during their hydrolysis during soil weathering. Aluminium, however, dominates 
because of its abundance. 



2.3 ALKALI AND ALKALINE EARTH CATIONS 



The major alkali and alkaline earth cations — Na, K, Ca, and Mg — are roughly 2% 
try mass in igneous rocks. They are prominent in soils even though large amounts of 
these ions are lost during weathering of rocks to soils. The other alkali and alkaline 
earth metals — Li, Cs, Rb, Fr, Sr, Ba, and Ra — are present in only trace amounts, 
mg/kg and less, in rocks and soils. Although Be is in the alkaline earth column, it is 
omitted here because it reacts chemically like aluminium. In sedimentary rocks, the 
relative amounts are probably Ca > Mg = K » Na because weathering has released 
many ions before the rocks form. Na is released to the soil solution and leaches from 
the soil to the sea. 

The alkali and alkaline earth cations are the major cations in the soil solution, as 
aquated (water-surrounded) ions and as the charge-neutralizing cations on the sur- 
faces of soil colloids. The ions associated with soil surfaces can easily be exchanged 



36 IMPORTANT IONS 

for other cations. If a soil suspension contained initially equal amounts of the four 
major cations, their distribution would be Na > K > Mg > Ca in the salts in the wa- 
ter phase away from colloidal surfaces (the bulk solution) and Ca > Mg > K > Na 
in the ions associated with clay surfaces and organic matter (the exchanger phase). 
The Ca > Mg > K > Na relation is also the relative strength by which soil colloids 
retain the cations. In soils in humid and temperate regions, most of the cations are as- 
sociated with colloids, so the amount of cations exceeds the amount of water-soluble 
anions. 

The cations in productive agricultural soils are present in the order Ca 2+ > 
Mg 2+ > K + > Na + . Deviations from this order can create ion-imbalance problems 
for plants. High Mg, for example, can occur in soils formed from basaltic serpentine 
rocks. The Mg inhibits Ca uptake by plants. High Na occurs in soils where water 
drainage is poor and evaporation rates exceed rainfall. High Na creates problems of 
low water flow and availability in soils. Low Ca occurs in acid soils. 

The total salt concentration in the bulk solution of well-drained soils from humid 
and temperate regions is generally in the range of 0.001 to 0.01 M. In irrigated and 
arid soils, the soluble salt concentration is higher. It may be five to ten times higher 
than that of the irrigation water applied, because evapotranspiration leaves the salts 
behind. Where salts accumulate due to improper irrigation, high groundwater ta- 
bles, or seawater intrusion, salt concentrations (particularly Na salts) can reach 0.1 
to 0.5 M. 

The trace alkali and alkaline earth cations are present in the following amounts: 
lithium, 10-300 mg kg" 1 ; rubidium, 20-500 mg kg" 1 ; beryllium, 0.5-10 mg kg"'; 
strontium, 600-1000 mg kg" ; barium, 100-3000 mg kg; and radium, perhaps 
10~ 7 mg kg. Some varieties of fruit trees are sensitive to as little as 1 mg L" 1 
Li + in irrigation water, but Li + toxicity is rare. Rubidium, cesium, strontium, and 
barium have all been studied in the laboratory, but have received little attention in the 
field. Strontium has been studied because its radioactive isotope 90 Sr (half-life = 28 
years) is produced by nuclear fission and could cause long-term soil contamination 
after nuclear explosions or accidents. In soils the toxic Be 2+ ion behaves more like 
Al 3+ than like the other alkaline earth cations. 

The strength of cation retention by soil particles increases with increasing ion 
charge and with decreasing hydrated ion size. For the monovalent cations, the in- 
creasing order of retention is 

Li < Na < NH4 = K < Rb < Cs < Ag < H(A1) 

The H(A1) indicates that the Al 3 ' 1 " ion is responsible for the acidity in acid soils; 
the amounts of H + are quite small. Only Na, NH4, K, and H(A1) are in significant 
amounts in natural soils. 

The increasing order of retention of divalent cations by soils is 

Mg < Ca < Sr < Ba 

Only Mg and Ca are common in soils, and Ca dominates the cations associated with 
clay surfaces and organic matter in most soils. 



ALKALI AND ALKALINE EARTH CATIONS 37 



2.3.1 Calcium 



Most economic crops yield best in soils when Ca 2+ dominates the exchangeable 
cations. High Ca indicates a near-neutral pH, which is desirable for most plants and 
soil microorganisms. Calcium is an essential element for plants and animals, and the 
amounts are rarely deficient in soils. Other problems appear before the Ca itself is 
deficient: Mg in soils derived from Mg-rich serpentine rocks competes with Ca for 
plant uptake, and the Al 3+ in strongly acidic soils can severely hinder plant growth 
even though the amount of available Ca is still adequate. When the exchangeable 
Na + concentration exceeds 5-15%, water flow problems in the soil can result. 

The goal of reclaiming problem soils is to replace most of the exchangeable Al or 
Na with Ca so that the surface is dominated by Ca. Limestone (CaC03> neutralizes 
soil acidity and supplies Ca in acid soils. Gypsum (CaSC^ 2H2O) and other materials 
supply Ca to displace the Na in sodic soils. These treatments have to be repeated 
periodically as weathering leaches Ca from the soil in humid regions and irrigation 
water adds Na in arid regions. Both maintenance treatments yield long-term rather 
than immediate economic returns. Since farmers are understandably concerned about 
the short term, they want to delay these relatively expensive treatments as long as 
possible. Assessing how much and when treatment is needed is an important concern 
for soil chemistry. 

The process of weathering includes the continual loss of the alkali and alka- 
line earth elements from soils. The loss of Ca 2+ following liming of acid soils is 
about equal to the rate of loss during natural soil weathering. The average rate of Ca 
loss from natural weathering is approximately 10 4 moles Ca ha -1 yr" 1 , 400 kg Ca 
ha -1 yr -1 , or 1 tonne limestone ha -1 yr -1 . 

Tn the subsoils of arid and semiarid soils, Ca commonly precipitates as calcite 
(CaC03) rather than being leached away. It is found as indurated layers (caliche and 
other local names) in many arid soils and as more diffuse CaC03 in Aridisols and 
Mollisols. Precipitation of CaCC>3 in soils is affected by the rates of soil water move- 
ment, CO9 production by roots and microbes, CO2 diffusion to the atmosphere, and 
water loss by soil evaporation and plant transpiration. CaCC>3 layers are also derived 
from upward movement and evaporation of Ca-rich waters. Calcium carbonate ac- 
cumulations can amount to as much as 90% of the mass of affected soil horizons. 
Gypsum precipitates in some arid soils, despite being about lOx as water soluble as 
Ca carbonate. 

Although Ca is very important in plant nutrition, soils derived from limestone 
(crude CaCOs) parent material can be unproductive. The weathering of limestone 
releases only Ca 2+ and HCOJj", which leach away, leaving no soil behind. An exam- 
ple is the Terra Rossa soils of Italy. The limestone parent material lacks silicates to 
form the cation exchange capacity that retains Ca and other cations. The fertility of 
limestone-derived soils generally increases with the amount of silicate impurities in 
the parent material. Organic limestone soils, such as the peat soils in central Florida, 
managed like sand cultures, can be highly productive because managing their nutri- 
ent additions is relatively easy. They require high fertilizer inputs and the soils are 
unable to retain them effectively. 



38 IMPORTANT IONS 

2.3.2 Magnesium 

Despite being the second most abundant exchangeable cation in soils, Mg 2+ is the 
least-studied ion in this group. Excessive or deficient amounts are uncommon. Plant 
Mg deficiencies have been reported in some acid sandy soils and are a concern in 
northern Europe as a by-product of the large amounts of acid rain during the past 
century and continuing. Liming often corrects both the acidity and a Mg 2+ defi- 
ciency, because agricultural limestone usually contains appreciable Mg impurities. 
Under chronic conditions and with crops having high Mg requirements, dolomite 
(CaMg(C03)2) or limestone containing dolomite is a satisfactory liming material. 

High exchangeable Mg is sometimes associated with low water permeability, soil 
crusting, and high pH, similar to the characteristic conditions of sodic (Na-rich) soils. 
This is sometimes the result of soil formation under marine conditions, where Na' 1 " 
and Mg 2+ predominate. The Na + may have produced the poor soil structure, leading 
to low water permeability, and then leached away, leaving a Mg soil with an inherited 
soil structure. Serpentine (an Mg silicate rock)-derived soils have high Mg 2+ levels 
that repress Ca availability to plants. 

Magnesium is an important constituent of many primary and secondary alu- 
minosilicate minerals (with the exception of the feldspars). Magnesium in mafic 
(Mg 2+ - and Fe 2+ -rich) minerals often leads to the formation of chlorite and mont- 
morillonite clay minerals in soils. 



2.3.3 Potassium (Kalium) 

Potassium is the third most important fertilizer element, in terms of amounts added as 
fertilizer, after nitrogen and phosphorus. Many soils of humid and temperate regions 
are unable to supply sufficient K + for agronomic crops. Farmers in these areas long 
ago recognized the benefits of applying wood ash and other liming materials to their 
acid soils. Both the alkalinity of the ash (to counter Al 3+ toxicity) and its K and Ca 
content are beneficial. 

Soils retain K + more strongly than Na + , because the hydrated K ion is smaller 
than the hydrated Na ion. In addition, K fits well between the sheets of several soil 
clay minerals, while Na does not, so K is retained strongly in soils containing these 
clay minerals. The K concentration in soil solutions is low but is replenished by K 
diffusion from between the sheets of these clay minerals, from the slower weathering 
of K-containing feldspar minerals, and from the decay of soil organic matter. 

The K concentration in the soil solution must be continuously replenished to sup- 
ply plants. Soil testing methods using extracting solutions try to measure this replen- 
ishment rate. The amount of K extracted depends on the ability of the extracting 
cation to separate the silicate sheets and the ability of the anion to react with K + . 
Sodium tetraphenylborate is a strong K + extractant because K + forms a complex ion 
with the tetraphenylborate ion and because Na + tends to separate the silicate sheets. 
The exchangeable plus soluble K content determined by this single-extraction soil 
test is often closely correlated to the amount of K that the soil will supply to plants 
over a growing season. 



ALKALI AND ALKALINE EARTH CATIONS 39 

The ability of some soils to supply K + for plants is remarkable. Tropical soils 
derived from easily weathered basaltic rocks have supplied up to 250 kg ha~ ! yr _1 
of K to banana plants for many years without noticeable soil impoverishment. Such 
rocks typically contain high concentrations of K-containing minerals, which weather 
rapidly because of their small crystal size and the climate. Soils in temperate regions 
that supply adequate K for crop needs often contain considerable K-containing mica 
in their clay fractions. 

Continual K fertilization is necessary for K-deficient soils. In addition to inher- 
ently low total K in such soils, certain layer silicates strongly retain, or fix, much 
of the added fertilizer K. Fixed K is released back to the soil solution too slowly to 
satisfy plant needs. Soils containing appreciable sand- and silt-sized vermiculite are 
particularly troublesome in this regard. Saturating the K-fixation sites is uneconom- 
ical because of the large amounts of K required. Potassium losses by leaching are 
small compared to the amounts of K fixation, except in very sandy soils. Estimates 
for sources of K fertilizer indicate a 800-yr lifetime at present rates of use. 

2.3.4 Sodium (Natrium) 

Sodium, in contrast to K + , is a soil chemical concern when it occurs in excess, more 
than 5-15% of the exchangeable cations. Sodium can accumulate in these amounts in 
areas inundated by seawater, in arid areas where salts naturally accumulate from the 
evaporation of incoming surface or ground water, and in irrigated soils because irri- 
gation water often contains high Na. The 5-15% exchangeable Na + can inhibit wa- 
ter movement into and through many soils. The lower value applies to fine-textured 
soils, especially those containing high contents of swelling clays, and soils wetted 
by rainwater. A high exchangeable Na content can work to the benefit of water in- 
filtration, on the other hand, for strongly swelling soils if the extensive cracks that 
develop upon drying allow water penetration. Also, sandy soils can be irrigated at 20 
or 30%; exchangeable sodium. The high Na status can slow water infiltration to more 
manageable rates during irrigation. Chapter 1 1 is largely devoted to Na problems in 
soils. 

Sodium is not required by plants but can replace part of the K requirement of some 
plant species. The attraction of some animals to NaCl and the long history of NaCl 
as an important article of commerce suggest that soils in humid regions may provide 
inadequate Na + and Cl~ for animal diets. Humans ingest more NaCl than they re- 
quire, because of taste preferences and because NaCl is used in food preservation, 
but a suggestion of insufficiency in animal diets remains. Of the essential elements, 
Na + , CI - , l~, and F~ are unique in that much of their supply to humans comes from 
additions to our diet rather than from natural foodstuffs. 

Saline soils are a problem for plants because the high osmotic potential of the soil 
solution makes it unavailable for plants. The plant has to expend so much energy to 
take up water that little energy is left for growth and crop yield. This is similar to the 
problem of organisms in marine water — "all that water and none of it fit to drink." 
Sodium is toxic to some plants at high concentrations, but for most plants this is a 
relatively minor problem compared with the restricted water uptake and movement 



40 IMPORTANT IONS 

that normally precede Na toxicity. Fruit and nut trees and berries are sensitive to 
Na and may show toxicity symptoms before water deprivation. The high soil pH 
that accompanies Na accumulation in arid soils is generally of secondary importance 
compared to water problems and the microelement deficiencies induced by the high 
pH. In cold regions, the amounts of NaCl added to roads for snow and ice clearance 
seem to be too small to cause significant plant damage, because of dilution by snow 
and rain. 

Most plant roots repel most of the NaCl in the soil solution. Halophytes are dis- 
tinctive because they allow NaCl to enter, which is a great advantage in high-salt 
soils and waters. Research to adapt these plants to agricultural use is under way. The 
problems are that the seeds are too small to harvest easily and the plants are too salty 
to be edible forage for grazing animals. 

The. alkali and alkaline earth cations are (he major exchangeable cations, and they 
determine the pH of most soils. The relation of the dominant exchangeable cations 
to soil pH can be summarized by the following reactions: 

soil — Al 3+ + H 2 = soil— A10H 2+ + H + pH ~ 5 (2.1) 

soil— (Ca 2+ , Mg 2+ ) + H 2 = no net reaction pH ~ 7 (2.2) 

soil— Na + + H 2 = soil— H+ + Na + + OH" pH ~ 9 (2.3) 

The Ca 2+ - and Mg 2+ -saturated soils hydrolyze water very little, so these soils have 
a neutral pH. Exchangeable Na + , particularly at low salt concentrations, is weakly 
adsorbed by soil clay. This causes some water molecules to dissociate into the more 
strongly adsorbed H + and leave OFF in the soil solution. Acid soils characteristi- 
cally have significant amounts of exchangeable Al ?f . The amounts of H + in acid 
soils are very small. Neutralizing acid soils means neutralizing the large amounts 
of available Al, the reserve acidity, in the soil. Some soils are more acid than pH 5 
due to organic acids from decaying organic matter or due to the microbial oxidation 
of sulfur and sulfides to sulfuric acid. Other sources of acidity are phosphate and 
nitrogen fertilizers and acid rain. 

Although most plants and many soil microbes are hampered al soil acidities as 
low as pH 5, plants in water culture grow well down to pH 3. The effects of soil 
acidity are due to toxicity by Al + and transition metal ions that are more soluble at 
low pH and coincidentally release H + by reacting strongly with water. The pH is an 
indicator of ion status in soils rather than being harmful per se. 



2.4 MAJOR SOLUBLE ANIONS 

The anion concentration in the aqueous phase of most nonsaline soils is less than 
the cation concentration. Much of the negative charge in these soils is due to soil 
colloidal particles. Most soil colloids have a net negative charge, which is balanced 
by cations in the water surrounding the colloidal particles. Those conditions apply to 
North American and European soils, which are relatively young and weakly weath- 



MAJOR SOLUBLE ANIONS 41 

ered. Other continents, where weathering has been intensive, have soil clays with 
both appreciable positive and negative charge. Positive charge increases anion reten- 
tion, particularly sulfate, in the same way as the negative charge retains cations. 

The major anions in the soil solution are CF, HCOJ, SO 2 ^, and NOJ. The soil 
solution concentrations can indicate the sulfur and nitrogen availability for plants in 
those soils. In humid region soils, the anion sum rarely exceeds 0.0 1 M in the soil 
solution; in arid regions the concentration can reach 0.1 M. The relative amounts of 
these anions vary with fertilizer and management practices, mineralogy, microbial 
and higher-plant activity, saltwater encroachment, irrigation water composition, and 
atmospheric fallout. 

In saline soils, anion (and cation) concentrations are higher. A typical distribution 
in the soil solution is C\~ > SO4 > HCO^~ > NOJ. At high pH (pH > 8.5), the 
distribution might be (HCOJ + CO, - ) > Cl~ > SO 2- > NO". 

The major soluble anions are retained weakly by most soils. Nitrate and Cl~ move 
through soils at virtually the same rate as the water. Sulfate and HCO^" lag slightly 
behind the wetting front because they interact with Ca 2+ , Mg 2+ , and Al 3+ and with 
positively charged sites on clay particles. This interaction is weak, however, com- 
pared to the strong retention of anions such as phosphate. In strongly weathered 
soils rich in Al and Fe(ni) hydroxyoxides, anion retention of sulfate and phosphate 
increases greatly. These soils can develop significant positive charge, so their anion 
retention (exchange) capacity can exceed their cation retention (exchange) capacity. 

The nitrate concentration in aerobic soil solutions is an index of, but only a small 
fraction of, total soil nitrogen. The nitrate concentration reflects a steady state rep- 
resenting nitrogen turnover in the soil and plant availability of nitrogen in the soil. 
Fertilization can temporarily change this steady state until denitrification, leaching, 
and nitrogen uptake by plants and microbes restore the nitrogen balance. 

Nitrate is actively taken up and reduced by soil organisms, but is chemically inert 
in the absence of soil microbes. If leached below the surface soil horizons and root 
zone, nitrate tends to move unhindered and unchanged through the subsoil. Unwise 
fertilization and organic waste disposal can therefore increase the NO3 concentra- 
tions in groundwaters and drainage waters. If downward water flow is very slow and 
the groundwater table is deep, as in arid regions, the time may allow nitrate to slowly 
disappear by microbial transformations. 

The contribution of agricultural practices to nitrate pollution of streams and well 
waters is a hot issue. Agriculture increases the nitrogen concentrations of these wa- 
ters, but in some cases land clearing and leaching of nitrogen from geologic strata 
in arid regions where nitrate had accumulated before cropping contribute to nitrate 
in the groundwater. Nitrogen in soil organic matter eroded from soils subsequently 
oxidizes to NOJ in aerated streams. 

Nitrate is the major anion of pollution concern because of its effects on the ecol- 
ogy of streams and lakes and because of potential harm to infants. The current upper 
limit of NO J in drinking water in the United States is 1 mg L~ ' nitrate as N (45 mg 
L - ' as NO3) or 7 x 10~ 5 M. The nitrate is reduced to nitrite (NO^) in the body, but 
infants lack the enzymes to reduce N07 further. Infants can accumulate toxic NOT 



42 IMPORTANT IONS 

concentrations if their water or diet contains high NO^~ concentrations. The nitrate 
concentration in the soil solution is greatest near the soil surface, where the ion is 
produced by microbial decay of organic matter, but it can leach rapidly. Root and 
microbial uptake reduce the nitrate concentration in the root zone. 

The major sources of soil NO^ are nitrate fertilizers and the oxidation of organic 
and fertilizer nitrogen. Rainfall contributes an additional small amount, perhaps 1 to 
2 kg ha -1 yr _1 , or about 1% of the nitrogen turnover. The first raindrops in a storm 
front contain NOJ from natural and anthropogenic NO., washed out of the air, plus 
HNO3 formed by lightning. Direct absorption of NO., (NO plus NO2-N2O4) gases 
from the atmosphere in industrial regions may add an additional 1 to 3 x 10 10 kg 
yr -1 (2 to 4 kg ha -1 yr -1 ) of N to soils. 

Nitrogen and, to a slower extent, sulfur cycle between the soil and the atmo- 
sphere. Acid rain in the atmosphere contributes considerable sulfur to soils. Depend- 
ing on the proximity to industrial sources, the sulfur fallout ranges from to 100 
kg ha - 1 yr - ' . In humid regions the highest rates are accompanied by plant damage 
from SO2 and H2SO4, and perhaps by higher soil acidity, which increases Al and 
transition meta] toxicity to roots. The lower rates of sulfur fallout can be beneficial 
to plants in regions where the native soil sulfur content is marginal to low. The bene- 
fits are enhanced if the atmospheric acidity has already been neutralized by dust and 
NH^J" before the sulfur contacts plants and the soil. 

The sulfate concentrations in soil solutions are good indicators of sulfur avail- 
ability to plants. Sulfur is being increasingly recognized as in short supply in humid 
region soils. The traditional NPK fertilizer scheme is being broadened to NPKS. 
The extent of sulfur deficiency in soils is increasing as superphosphate fertilizer is 
being replaced by triple superphosphate. Superphosphate contains 0.5 mol fraction 
sulfate, while triple superphosphate contains none. The sulfate supply in arid soils is 
adequate for plant needs. 

Sulfur readily changes oxidation state in soils between sulfate (SO^ - ), elemental 
S, and sulfide (S 2- ) as oxidizing-reduction conditions change. Several species of 
sulfur bacteria catalyze these changes. Sulfide reacts strongly with Fe and other tran- 
sition metals in soils so H2S evolution from soils is minimal. Any H2S formed during 
decay of organic matter diffuses through soil pores before reaching the atmosphere. 
Diffusion is slow, leaving ample time to react with the soil's solid phase. 

Sulfate ions are retained rather strongly in acid soils, in soils having appreciable 
positively charged clays, and in soils having considerable Al and Fe hydroxyoxides. 
The sulfate behaves somewhat as if AIOHSO4 is formed, although this compound 
has not been identified in soils. 

2.4.1 Halides 

The monovalent anions F - , CI - , Br - , and I - are the only oxidation states of the 
halogens in soils. Chloride is essential for plants in trace amounts and for animals in 
larger amounts. Although F is held to some degree in soils, as a group these ions are 
retained weakly and are in low amounts in well-drained soils. Plants are much more 
tolerant of high CI concentrations than of high concentrations of other micronutrient 



MAJOR SOLUBLE ANIONS 43 

ions. Except for the specific CI sensitivity of fruil and nut trees and berries, the effect 
of excess CI in soils is mainly to increase the osmotic pressure of soil water and 
thereby to reduce water availability to plants. Plants absorb CI~ from the soil solution 
and the biomass is a significant CI reservoir in terrestrial environments. As mentioned 
for Na + , the natural rate of CI supply to animals from soils through plants may be 
less than optimal. 

The loss of CI from the continents to the sea is about 3 x 10 12 mol yr _1 (or 
about 200 mol ha -1 yr~' or 8 kg ha -1 yr _1 ). Assuming that the oceans are a steady 
state, this equals the input rate to soils from the atmosphere and from parent material 
weathering. Chloride is only a minor constituent of igneous rocks, although apatite 
contains up to 0.0 1 mol fraction of CI and micas can contain slight amounts. Most 
of the CI input to soils is from rain, marine aerosols, salts trapped in soil parent 
materials of marine origin, and volcanic emissions. Chloride fallout at seacoasts can 
be as much as 100 kg ha" 1 yr~', but this rate decreases rapidly with distance to 1 
or 2 kg ha -1 yr~ ' in the continental interiors. This is apparently adequate for plants, 
since natural CI deficiencies in plants have not yet been found. In recent years water 
softening, industrial brines, and road deicing have contributed CI to local areas. 

Except for much greater CI accumulation in soils of arid regions, the soil chem- 
istry of I and Br resembles that of CI, except that I and Br are retained more strongly, 
especially by acid soils. The major input of I to soils appears to be atmospheric. En- 
demic iodine deficiency (goiter in humans) occurs in mountainous and continental 
areas isolated from the sea. Fortunately, supplementing NaCl with small amounts of 
I effectively supplies the I required in animal diets. Iodide and Br are both potentially 
toxic, but no natural cases have been reported. Bromide reactions in soils have been 
investigated as a tracer for the movement of water, nitrate, and soil solutions in soils. 

Fluoride is the most unique halide chemically and is the most common halide in 
igneous rocks. The igneous minerals fluorspar (CaF2) and apatite (Ca5(F,OH)(P04)3) 
are both insoluble in water. Fluoride can substitute for OH - to some extent in soil 
minerals. This mechanism is probably also responsible for F retention by aluminium 
and iron hydroxides in acid soils. Fluoride also associates strongly with H + . HF is a 
weak acid, pA" = 3.45. 

Fluoride concentrations in soil solutions, groundwaters, and surface waters of hu- 
mid and temperate regions are generally less than 5 x 10 -6 M (0. 1 mg L~ ' ). In arid 
regions, the F~ concentrations in groundwaters can reach 10~ 4 M (2 mg L _1 ) and 
higher. Some well waters near Pecos, Texas and Phoenix, Arizona contain as much as 
10~ 3 M (20 mg L - ') F. United States public health agencies recommend that drink- 
ing water contain about 2-5 x 10~ 5 M F (0.5- L mg L~ ' ) to reduce dental caries. The 
amount of F in our diet has increased due to increased phosphate fertilization. The F 
impurities in the fertilizer are absorbed to some extent by plants. 

Possible fluoride air pollution has aroused concern near phosphate fertilizer plants 
and aluminium and iron smelters. The F2, HF, and SiF4 gaseous by-products of these 
industries are potentially toxic to plants and animals. Once in the soil, however, fluo- 
ride is considerably less hazardous. It is retained somewhat by all soils. Soils receiv- 
ing high amounts of phosphate fertilizer may acquire as much as 20 kg ha -1 yr _1 of 
fluorine impurities from the fertilizer. This input is apparently harmless. 



44 



IMPORTANT IONS 



Table 2.2. Average content of oxyanions in the lithosphere and 
ranges in soils 





Average in 


Range 




Lithosphere 


in Soils 


Oxyanion 


(mgkg -1 ) 


(mgkg -1 ) 


B 3 + (BO 3 , - ) 


10 


2-100 


N 5+ (NOp 


Negligible 


100-1 000 


C 4+ (CO 2- ) 


— 


100-50000 


Si 4 + (SiO 4 ") 


250000 


100000-400000 


P 5+ (PO 3 -) 
S 6+ (SO;p 


1000 


100-1000 


500 


200-10000 


As 3+ (AsO 3- ) 


2 


0.1-40 


Se 4+ (SeO 2- ) 


0.1 


0.03-2 



2.5 WEAKLY SOLUBLE ANIONS 

The weakly soluble anions are primarily oxyanions (H3BO3, FLjSiCU, H2PO4, 
HM0O4 , SeO^ - , and H2As0 4 ~ are the forms at pH 6-7) — small, highly charged 
cations surrounded by strongly associated oxygen or hydroxyl ions. The average 
contents in soils and in the lithosphere are shown in Table 2.2. Selenium behaves in 
some respects like sulfate but is included in this group because of its low concentra- 
tions in soils. The more soluble sulfate oxyanions are discussed above. 

The weakly soluble oxyanions are weak acids; in aqueous solution they gain and 
lose H + as the acidity increases and decreases. In solids, the number of H + ions de- 
pends on the crystal structure and the need for charge neutralization in the structure. 
The oxyanions lend to be weakly soluble in the soil solution because they associate 
strongly with Fe 3+ , Al 3+ , Ca 2+ , Mg 2+ , and transition metal ions to form uncharged, 
and therefore insoluble, molecules. Phosphate is the weakly soluble anion of greatest 
economic, agricultural, and political interest, so it is discussed in a separate section. 



2.5.1 Boron, Silicon, Molybdenum, Arsenic, and Selenium 

The major soil species by far is silicate, which is the backbone of soil and rock min- 
eral structures. The soluble silica (H4Si04, or better described as Si(OH) 4 ) concen- 
tration in soil solutions ranges from about 6-10 mg L -1 (2 x 10 -4 M) in temperate 
region soils to 100 mg L _l (2 x 10 -3 M) in high pH soils and soils containing amor- 
phous silica and opal. Silicon is necessary to plants and animals in trace amounts and 
is very beneficial to plants. Silica lodges in cell walls and provides physical strength 
and resistance to insect and fungal attack. Silica may also aid in overcoming plant in- 
jury from soil salinity. These benefits are more evident in hydroponics because many 
growth solutions lack silicon. 

Si(OH) 4 is a very weak acid, pK\ = 9. 1 . It polymerizes and precipitates as amor- 
phous silica when concentrations reach 10 -3 M. Such high soluble silica concentra- 



WEAKLY SOLUBLE ANIONS 45 

lions probably occur only in basic soil solutions, in the interstitial solution between 
expanded 2: 1 layer lattice silicates, and perhaps on the surfaces of actively weath- 
ering minerals. Concentrations much below 10 -4 M in soil solutions are unusual in 
soils. 

Evolution has taken little advantage of silicon's ubiquity and relatively constant 
solubility. Silicon is a useful strengthening agent of the plant's cell wall and forms 
a cast of the cell wall's morphology. These can remain intact in soil as "phytoliths" 
after the plant decays. Silicon is essential for animals and only in trace amounts. 

Silicon reactions are central to rock weathering and soil development. Silicon is 
the soil component lost in greatest amount from rock minerals during weathering, 
and the transformations of silica into secondary minerals are the major reactions 
of soil development. The sand fraction of soils is usually >90% quartz (SiCh), the 
most prevalent form of Si in soils. Highly weathered soils may contain as little as 
20% Si (Table 2.1a). Al and Fe ore deposits are essentially highly weathered soils 
from which most of the Si has been lost. 

Secondary silicates form as clay minerals in soils after weathering of the pri- 
mary silicates in igneous minerals. The secondary silicates include amorphous silica 
(opal) al high soluble silica concentrations and the very important aluminosilicate 
clay minerals: kaolinite, smectite (montmorillonite), vermiculite, hydrous mica (il- 
lite), and others. Kaolinite tends to form at the low silicate concentrations of humid 
soils, whereas smectite forms at the higher silicate and Ca concentrations of arid 
and semiarid soils. The clay fraction of soils usually contains a mixture of these 
clay minerals, plus considerable amorphous silicate material, such as allophane and 
imogolite, which may not be identifiable by x-ray diffraction. 

Borates, molybdates, selenates, and arsenates occur in such trace quantities in 
soils that they probably exist only as impurities in major soil particles and on particle 
surfaces rather than as separate minerals. Soils in humid regions sometimes benefit 
from borate and, less often, molybdate additions. The range between deficient and 
excess is narrow, so spreading a few kg ha -1 (a few lbs/acre) uniformly over the soil 
is difficult unless the borate and molybdate salts are mixed in with other fertilizers 
or inert materials. 

Boron is widely and rather uniformly distributed in rocks and sediments. One 
mineral is tourmaline ((Mg,Fe,Ca,Na,Li,K)4-8B2AbSi402o(OH)2), but more com- 
monly B is a minor impurity in other minerals. In soils B is diffuse and is not in 
identifiable B minerals. Marine sediments were once thought to be characteristically 
higher in boron than terrestrial sediments, but this view is no longer so widely ac- 
cepted. Boron released to solution by weathering interacts primarily with Fe and Al 
hydroxyoxides, with maximum adsorption at pH 7 to 9. Aluminosilicates adsorb B 
only weakly. 

Boron exists in solution as boric acid (H3BO3 or B(OH)3, pK = 9.2). Higher 
borate polymers such as borax (Na2B4C>7 • IOH2O), the common boron fertilizer, 
dissociate to monomers in dilute solutions. Plant deficiencies of boron are well doc- 
umented in highly weathered soils, and boron toxicity is known in arid and irrigated 
soils. The range between deficiency and toxicity in soils is narrower for boron than 
for any other essential element. Boron concentrations greater than a few milligrams 



46 IMPORTANT IONS 

per liter in the soil solution can be toxic to sensitive plants, and concentrations less 
than several tenths of a milligram per liter may indicate deficiency. 

Borates are slightly volatile. Evaporation and condensation from the atmosphere 
may circulate boron geochemically and may have contributed to a rather uniform dis- 
tribution ofB in soils worldwide. Boron's ubiquitous nature and buffering of boron 
concentrations by soils has apparently allowed life to evolve with a narrow range of 
boron sufficiency. 

Molybdenum is present in soil solutions as molybdate (H2M0O4, pK «* 5). 
Molybdate can reduce to MoOj under reducing conditions. As is true for boron, 
molybdate forms complex polymeric ions at high concentrations in water. In soil 
solutions, however, only the monomer exists. 

Molybdate reacts strongly with Fe hydroxyoxides. Its solubility and plant avail- 
ability increase with increasing pH. The soluble Mo concentration may be quite low 
in acid soils but reaches an upper limit as pH increases. Soil solutions at pH 6.5 may 
contain about 3 x 10 -8 M Mo (3 /xg Mo L _1 ). Groundwater concentrations greater 
than 5 x 10~ 8 M (5 /tig molybdenum L _1 ) are unusual. 

Molybdenum is essential for the symbiotic nitrogen-fixing microorganisms grow- 
ing on the root nodules of legumes and some other plants. Mo deficiency occurred 
in the Victoria province of Australia. Mo fertilization at a rate of only 70 g ha - ' in- 
creased forage production 12-16 times by stimulating nitrogen fixation in legumes. 
In other areas of the world, Mo deficiency in some acid soils can result from ad- 
equate soil content but inadequate availability. Liming corrects the acidity and the 
low Mo availability. Direct Mo fertilization is also effective, although Mo salts, as 
with boron, must be diluted and spread carefully to prevent Mo toxicity. 

Arsenic is an essential microelement for animals, but As is known mainly as a 
toxic element. Arsenic problems in soils are primarily the result of anthropogenic 
activities. Lead arsenate and copper acetate-arsenate ("Paris green") were once com- 
mon insecticides. Spraying apple trees in Washington with Paris green for many 
years led to As concentrations high enough to harm the trees and hinder replanting. 
The soils have recovered after several decades as the Pb and As in the soils have 
become less available. The recovery is not due to Pb and As leaching below the root 
zone. Movement of As through soils is minimal unless large quantities are concen- 
trated in a small or unreactive soil volume, such as might be the case for industrial 
waste disposal on coarse-textured soils. 

The soil chemistry of arsenate (AsO^ ) resembles that of phosphate. Arsenate, 
however, can be reduced to arsenite. Arsenate is considered to be the state existing 
in soils, but As concentrations of soil solutions increase under reducing conditions, 
suggesting reduction to arsenite ( AsO^ , Eh° = +0.56 v). Both H3 ASO4 and HASO2 
are weak acids and are strongly retained by soils. Elemental As, arsine (ASH3), and 
AS2S3 are stable under strongly reducing conditions, but whether they form to an 
appreciable extent in soils is unknown. As with other cations and anions that are 
weakly soluble in soils, arsenic added to the soil solution decreases in concentration 
with time. The decrease is due to increased strength of retention rather than loss 
by volatilization of AsFTj. Arsenic is retained strongly in soil under oxidizing and 
reducing conditions. 



Selenium and sulfur chemistry are similar, except that selenates (Se0 4 ) and se- 
lenites (SeO^ - ) are weaker acids than their sulfur counterparts- Because of their 
weak-acid character, SeO;j - and SeOj~ should be retained more strongly by soils 
than the corresponding sulfur anions. The potential of the selenate-selenite couple, 
Eh° = 1.15 v, indicates that selenite is the stable oxidation state over most of the 
range of soil conditions, but selenate is the form most often reported. Selenate is 
stable under strongly oxidizing conditions, and elemental selenium is stable under 
reducing conditions (Eh = 0.74 - 0.059 logCSeOj - ) for the selenite-selenium cou- 
ple). Selenium redox reactions are apparently rather reversible and may not require 
microbial catalysis. Some reports indicated that Se may be lost from soils by reduc- 
tion to H2SC a gas. The amounts of F^Se evolved from soil, like H2S, are small. 

Selenium is essential for animals; natural Se deficiencies as well as toxicities have 
been recognized in grazing animals and fowls. Selenium salts may accumulate in 
poorly drained areas of arid regions, particularly in the northern high plains of North 
America. Legumes growing in these soils can accumulate toxic quantities of Se. Such 
plants are among the "loco weeds," which cause locomotion problems and eventual 
death of grazing animals. 

Plants from some areas in northern California have been found to be low in Se and 
animals benefit from Se supplements. This may be the basis for the current popularity 
of Se as a diet supplement for humans. Selenium gained notoriety in central Califor- 
nia when drainage water from a laTge irrigation project evaporated in a reservoir that 
is also a waterfowl nesting area. The evaporation concentrated Se and it was blamed 
for abnormalities in ducklings hatched there. The drainage water was originally sup- 
posed to drain to San Francisco Bay, where the Se might have been beneficial or at 
least be diluted to insignificance. Funding difficulties instead stopped the drainage 
line at the reservoir, which has no external drainage. The fear arose that Se might 
also be accumulating excessively in other poorly drained areas in the western United 
States and Canada, but research indicates that the California case is unique because 
the water originated from a region of high-Se soils. 



2.5.2 Phosphate 

Crops growing in large areas of the world's agricultural soils respond to phosphate 
fertilization. Phosphate is in the shortest supply of the major (NPKS) fertilizer ele- 
ments. One estimate is that present ore deposits can supply phosphate for no more 
than 200 years at current rates of use. Other phosphate deposits exist but they are of 
lower quality. 

Phosphate chemistry in soils has been studied more intensively than that of any 
other element save nitrogen. Phosphate added to soils is first adsorbed quickly and 
is later "fixed" into increasingly less soluble states as time increases. Despite this 
great effort, quantitative predictions of phosphate concentrations in soil solutions are 
poor and no techniques have been devised to release the large amounts of unavail- 
able phosphate in soils, nor to prevent fixation of fertilizer phosphate by soils. The 
uncertainties about soil phosphate chemistry and the difficulty of increasing phos- 



48 



IMPORTANT IONS 



phate availability are due to phosphate's strong interaction with many inorganic and 
organic soil components. The uptake by plants and microorganisms, continual return 
from organic decay, and slow phosphate release rates in soils continuously release 
phosphate but at rates that are often below those needed for optimal plant growth. 
Phosphate chemistry is of greatest interest in acid soils. Phosphate is increasingly 
unavailable as soil acidity increases, due to retention by Al and Fe. The decay of 
phosphate in organic matter bypasses the inorganic soil fraction to some extent and 
maintains some phosphate availability to plants over the whole range of soil pH. 

Phosphate is P's only oxidation state in soils and plants. Phosphite (POj~, phos- 
phine (PH3), or organophosphorus compounds of lower phosphorus (< V) oxidation 
states have not been found in soils. The H2PO4 and HPO^ ions are predominant at 
the pH of soil solutions. The pH distribution of the H3PO4-PO4" series in solu- 
tion is shown in Fig. 2.2. The phosphate mineral apatite (Ca 5 (OH,F)(P0 4 )3) is com- 
mon in rock minerals. In acid soils, most solid-phase phosphate is associated with 
Fe and Al and their hydroxyoxides. In basic soils, phosphate is associated with Ca 
in apatite-like forms. Whether phosphate truly adsorbs on surfaces or precipitates 
within the weathered layer on particle surfaces is uncertain. Phosphate fixation is ap- 
preciable in all but very coarse-textured soils and is particularly strong in soils rich 
in amorphous Fe and Al hydroxyoxides or allophane (in volcanic soils). Cycling of 
phosphate by reduction of Fe(III) phosphate during flooding and reoxidation after 
drainage is thought to account for the continued phosphate availability for centuries 
in paddy soils. 

Soil solution concentrations of phosphate are of the order 10 -6 to 10~ 7 M (0.01 
to 0. i mg V ' ). For the purpose of soil testing, many workers have tried to devise ex- 
traction procedures yielding phosphate values related to crop response of phosphate 



<oo 



H3PO4 



HjP0<f 



H PO4 



PO4 



< 

I 
a. 
in 
O 

1 

Q. 




FIGURE 2.2. The pH distribution of the phosphate series. 



WEAKLY SOLUBLE ANIONS 49 

fertilizer. Results from one area and one crop are often inapplicable elsewhere and to 
other crops. Radioactive 32 P enabled researchers to follow phosphate uptake and in- 
activation in soils. Soil phosphate was called labile or nonlabile, according to the rate 
at which it would exchange with radioactive phosphate. The labile fraction was con- 
sidered to be the amount available to plants. This attractive concept oversimplified 
the situation, however, and has lost favor. 

Later, researchers assumed equilibrium with various phosphate minerals to cal- 
culate phosphate ion concentrations in soil solutions. Minerals such as variscite 
(AIPO4 • 2H2O), strengite (FePC>4 ■ 2HbO), and the series between pure variscite 
and strengite have solubilities in the range of phosphate concentrations of acid soils. 
Soluble phosphate concentrations in basic soils are in the range of solubilities of 
octocalcium phosphate (Ca4H(P04)3). The correspondence, between soil concentra- 
tions and mineral solubility is unfortunately too crude to accurately predict the soil 
phosphate status for plants. 

Phosphate behavior is also described by Langmuir and Freundlich adsorption 
equations, although these models may be too simple to accurately explain soil 
phosphate behavior. Kinetic models of phosphate retention by soils are also being 
employed. Although kinetics can suggest retention mechanisms, the complexity of 
soil-phosphate behavior makes this prospect difficult to achieve. 

The calcium phosphate series ranges from the most soluble Ca(HPC>4)2 through 
CaHPC>4 and Ca4H(P04)3 (octocalcium phosphate) to the least soluble 
Ca5(OH,F)(P04)3 (apatite). Phosphate fertilizers are made by treating rock phos- 
phate, mostly apatite, with sulfuric acid to make superphosphate (nominally CaHP04) 
or with phosphoric acid to make triple superphosphate (nominally Ca(HP04>2)- The 
marketing hyperbole comes from the early use of rock phosphate as a fertilizer on 
acid soils. During the transformation of the fertilizers to Fe and Al phosphate in acid 
soils, or to octocalcium phosphate and apatite in alkaline soils, perhaps one-third 
of the phosphate becomes available to plants. The acid-treated phosphate is more 
soluble and hence more immediately available than rock phosphate. In the transi- 
tion from superphosphate to the "improved" triple superphosphate, crop yields in 
some cases declined. This was because superphosphate contains 50 mole percent 
sulfate, which overcame a sulfur deficiency in these soils. Triple superphosphate 
lacks sulfur. 

The phosphate concentrations of waters draining from soils usually are about 
10~ 7 M. Worldwide, this amounts to a phosphate loss of 17 x 10 10 moles yr _l 
(10 mol ha -1 yr~' or 1 kg ha' 1 yr _1 ). Phosphate in eroded soil particles reaching 
the sea is estimated to be an additional 13 x 10 10 mol yr~'. Fertilization affects the 
phosphate content of sediments eroded from surface soils, and increases the phos- 
phate concentrations of drainage waters and groundwaters. Preventing erosion has 
the added benefit of reducing phosphate inputs to streams and lakes. 

The rate of phosphate loss from soils by weathering is about the same as the over- 
all weathering rate, so the total amount of phosphate in soils tends to remain constant 
throughout soil development. The availability of phosphate to plants, however, de- 
creases as soils become more acid and the proportion of phosphorus as aluminium 
and iron phosphate increases. 



50 IMPORTANT IONS 

Human activities affect the global phosphorus cycle to an ever-increasing degree. 
The rate of phosphate mining equals or exceeds the rate of phosphate lost naturally 
from the continents. About 35 x 10 10 mol yr _1 are used as phosphate fertilizer, while 
6 x 10 10 mol yr _1 are used by other industries and in the home. The industrial forms 
are particularly soluble, so the phosphate in sewer discharge, as well as the lesser- 
available forms from farm runoff and erosion, is already affecting freshwater and 
marine ecology in North America, Europe, and Asia. 

Our ignorance of the state of phosphate in soils and our inability to increase the 
availability of soil and fertilizer phosphate ranks as one of the great frustrations 
and challenges of soil chemistry. The unsuccessful attempts to overcome soil fix- 
ation of phosphate fertilizer include (I) adding soluble silica to soils to replace ad- 
sorbed phosphate, (2) using ammonium phosphate fertilizers to avoid the strong acid- 
ity produced when calcium phosphates dissolve, (3) creating polymeric phosphates, 
which are more soluble than monomelic orthophosphates, in the hope of forming 
less readily fixed phosphate ions in soil solutions (soil microorganisms, however, 
rapidly break the phosphate-phosphate bonds of such polymers), and (4) using el- 
emental phosphorous, nitrogen-phosphorus, and organophosphate compounds. As 
yet, no economically or chemically feasible alternative to orthophosphate fertiliza- 
tion of soils has appeared, except for the centuries-old practice of increasing the pH 
of acid soils by liming. Phosphate availability from organic manures seems to be 
greater per unit of P, but their P contents are low. 



2.6 ALUMINIUM, HYDROGEN, AND TRANSITION METALS 

Aluminium in soils is closely connected to soil acidity and is also discussed in the 
chapters on acid soils and ion-water reactions. The acidity of acid soils is due to 
the reactions of water with exchangeable Al 3+ on the surface of soil particles. The 
strong Al-water reaction repels H + from the water molecules into the soil solution. 
This can create soil acidities as low as pH 4.5. Stronger acidity means other H + - 
yielding reactions — organic acids from soil organic matter decay, sulfur and sulfide 
oxidation, phosphate fertilizers, ammonia oxidation, acid rain, and Fe- and Mn- 
water reactions — are active. 

The Al-water reaction continuously and slowly liberates H + as Al in aluminosil- 
icate minerals weather to gibbsile (Al(OH)3>, the Al state in the most highly weath- 
ered soils. The Al-water reaction itself is fast, but the weathering of soil minerals 
to create surface Al 3+ is slow. Exchangeable Al 3+ and its hydrolyzed-polymerized 
forms (Al(OH),(H20)£_~' l)+ ) produce the acidity of most soils as they hydrolyze 
further toward Al(OH>3. The Al 3+ is tightly bound to clay surfaces; the exchange- 
ability of Al 3+ depends on the concentration and cation charge of the salts in the 
extracting solution. Most workers attribute low exchangeability to strong adsorp- 
tion of the polymeric ions on clay surfaces. Only A1(H20)^ + is considered truly 
exchangeable, and it is present in appreciable amounts in soils only at pH < 5.5. 
While other major exchangeable cations are leached from soils during weathering, 
Al is retained in soils as solid-phase Al(OH) 3 . The large amounts of limestone to 



ALUMINIUM, HYDROGEN, AND TRANSITION METALS 51 

neutralize ("sweeten") acid soils is due to the amounts of Al 3+ that have to convert 
toAl(OH) 3 . 

The amount of H + is an appreciable fraction of the total cations in the soil solu- 
tion only in extremely acid (pH < 4) soils. Such high concentrations are generally 
caused by active oxidation of sulfides and elemental sulfur, in the immediate vicin- 
ity of dissolving phosphate fertilizer granules, or from organic acids from decay of 
organic matter. High sulfide contents are typical of some mine wastes and recently 
drained coastal sediments, particularly in the tropics and subtropics. When the soils 
are drained and oxygen can enter the soil, sulfide microbially oxidizes to H2SO4, 
creating acidities as great as pH 1 .5 or 2. When the supply of sulfur is exhausted, the 
H2SO4 leaches away and the soil pH reverts to near-neutral values. When the soil is 
reflooded, the oxidation stops and the pH also returns to near neutrality as the sulfate 
is reduced to sulfur and sulfide. 

In mine spoils of humid regions, complete sulfur oxidation may require several 
years. During this time, plant growth is inhibited and erosion can be severe. The re- 
sulting soil damage can be evident for decades after acid production has ceased and 
the acidity has been leached away. In more arid regions, the natural reclamation of 
mine spoils, in terms of sulfide oxidation, proceeds even more slowly. Limited mois- 
ture inhibits sulfur oxidation and acidity production, as well as plant establishment. 
Plants are usually unaffected by the acidity because the basicity of arid region soils 
is adequate to neutralize the acidity produced. The salinity resulting from neutraliza- 
tion, however, can reduce plant growth. 

Although the H + concentration in soils is normally small, it is extremely im- 
portant. The chemistry of many ions and the activity of soil microorganisms is so 
closely tied to pH that no other single soil measurement conveys as much informa- 
tion. The measurement of soil pH in acid soils has received considerable study. The 
measurement is easy and seemingly straightforward but the simplicity is only appar- 
ent. Changes of the soil's electrical double layer, salt concentration, partial pressure 
of CO2, products of microbial reactions, and extraneous electrical potentials at the 
soil-electrode interface can cause significant aberrations in soil pH measurements. 
The pH measurement is closely tied to predictions of the lime requirement of acid 
soils. The pH of basic soils has less economic importance and has received corre- 
spondingly less attention. In addition, the pH of calcareous soils is often controlled 
by CaCC>3 dissolution. Laboratory pH measurements of calcareous soils, and field 
measurements at similar CO2 concentrations, tend to be monotonously similar, in the 
range from pH 7.5 to 8.5. A soil pH > 8,5 often indicates appreciable exchangeable 
Na + . High exchangeable sodium can also exist at lower pH under marine conditions. 

Transition metals (groups IB through VIIB and group VIII of the periodic table, 
Fig. 2.1) are distinct from the elements at either end of the periodic table in that 
electrons are added to and removed from inner electron orbitals. The chemistry of 
the transition metals therefore changes more subtly from element to element than 
elements having election changes only in the outer, .v and p, orbitals. In addition, 
many transition can have more than one oxidation state in soil. 

Under aerobic conditions, almost all of the transition metals plus aluminium and 
beryllium, essentially all of the metals in the periodic table except the alkali and al- 



52 IMPORTANT IONS 

kaline earth metals, exhibit one major reaction in soils. They associate strongly with 
O 2- and OH - ions and tend to precipitate as insoluble hydroxyoxides or as minor 
components of insoluble aluminosilicates. The precipitation may be on the surface 
and be called adsorption, but the effect on water solubility and plant availability is 
the same. The exceptions are Cd, Hg, Pb, Zn, and Cu, which also react extensively 
with organic matter and sulfide, if available. 

In anaerobic soils, the individual chemistry of the ions is more distinctive. The 
transition metal ions in the middle of each period of the periodic table — chromium, 
manganese, iron, nickel, cobalt, and copper — can reduce to lower oxidation states, 
while the end members — scandium, titanium, and zinc — have only one oxidation 
state. The lower oxidation states are more water soluble but still tend to precipi- 
tate as carbonates and sulfides, or associate with organic matter, thus reducing their 
movement but increasing their plant availability. 

Small amounts of transition metal and Al ions are associated with clay surfaces 
and soil organic matter. The amounts of exchangeable and water soluble ions increase 
with soil acidity. The approximate order of trivalent cation retention is 

Al < Fe < Sc < Y < Eu < Sm < Nd < Pr < Ce < La 

Only Al + is a common exchangeable cation in soils and is significant only in mod- 
erately to strongly acid soils, pH < 5.5. Al and Fe are usually the second and third 
most common cations in total soil content, but only tiny fractions are exchangeable 
on soil particle surfaces and dissolved in the soil solution. 

The important soil chemical elements of this group are aluminium and all (except 
scandium) of the first-row transition metals: titanium, vanadium, chromium, man- 
ganese, iron, cobalt, nickel, copper, and zinc. They are important either because their 
amounts in soils are large (Al, Fe, Ti) and therefore important to soil development, or 
because they are essential elements to living organisms. A few other transition metals 
outside of this first row — molybdenum, cadmium, tin, mercury, and lead — are also of 
interest because of their essentiality (Mo, Sn) or toxicity (Cd, Hg, Pb) to organisms. 
The total amounts in soils range from as much as 30% Fe and Al in highly weath- 
ered soils to less than 1 mg kg -1 (Table 2.3). Those not listed are normally present 
in concentrations of < 1 mg kg - ' . Trace contents in soils have been defined as < 1 00 
mg kg -1 . Plant uptake and decay, plus the strong retention of transition metals by 
inorganic and organic soil components, lead to slight accumulation of the trace metal 
ions near the soil surface. Total concentrations in the surface centimeters of unfilled 
soil can be several times the concentrations shown in Table 2.3, and several times 
those in their subsoils. 

The ions of this group precipitate from pure solutions (in the absence of soils) as 
hydroxyoxides of varying degrees of hydration, such as Al(OH) 3 , AlOOH, AI2O3, 
FeOOH, Fe2C>3, and MnO v . Their solubility products can all be expressed as 
(M)(OH)*. The ion concentrations in soil solutions and soil extracts of all ions 
in this group, except Fe(III) and aluminium, are usually several orders of magni- 
tude less than those calculated from their solubility products. This indicates that 
soils retain these ions much more strongly than do pure hydroxyoxides. The relative 
soil retention of these cations, however, roughly follows the order of the solubil- 



ALUMINIUM. HYDROGEN, AND TRANSITION METALS 53 

Table 2.3. Total contents of transition and related metal ions in 
the lithosphere and in soils a 





Average in 






Lithosphere 


Soil content 


Element 


(mg kg - ' ) 


(mgkg -1 ) 


Beryllium 


6 


1-40 


Aluminum 


70000 


10000-200000 


Titanium 


4400 


1000-10000 


Vanadium 


150 


20-500 


Chromium 


100 


5-3000 


Manganese 


1000 


200-3000 


Iron 


50000 


10000-300000 


Cobalt 


40 


1-50 


Nickel 


100 


10-1000 


Copper 


70 


2-100 


Zinc 


80 


10-300 


Yttrium 


30 


20-200 


Zirconium 


220 


60-2000 


Molybdenum 


2 


0.2-5 


Cadmium 


0.2 


0.01-7 


Tin 


40 


5-10 


Lanthanides, total 


— 


10-500 


Mercury 


40 


0.005-0.1 


Lead 


10 


2-200 



"After D. J. Swainc. 1955. Commonwealth Bureau of Soil Science Techni- 
cal Communications 48, Farnham Royal, Buckinghamshire. England; W. H. 
Fuller. 1977. EPA-600/2-77-020; and J. A. McKeague and M. S. Wolynetz. 
1980. Geodenna 24:299. 



ity products of the oxidized cations. Reducing conditions increase the Fe 24 " and 
Mn 2+ concentrations in the soil solution, and complexing by soluble organic anions 
increases the concentrations of Cu 2+ and Zn 2+ . 

Because these cations are multivalent, their hydroxyoxide ion products involve 
the second, third, or fourth power of the OH~ concentration. Their concentration 
changes in soil solutions, however, tend to be proportional to only the first power 
of OH~ or H + . This is partly explainable by the soluble ions being hydrolyzed. For 
example, the mechanism of Fe(III) dissolution and precipitation at the pH of normal 
soils is probably 

FeOOH + H + = Fe(OH)t (2.4) 

Hydrolysis explains only part of the disagreement between metal ion concentrations 
in soil solutions and corresponding solubility products. Soils apparently retain trace 
metal cations by other mechanisms in addition to precipitation as pure phases. Such 
mechanisms include coprecipitation as minor constituents in Fe, Al, Ti, Mn, and Si 
hydroxyoxides; adsorption on soil surfaces; complexation with organic matter; and 



54 IMPORTANT IONS 

incorporation into plant tissues and decay products. These mechanisms reduce cation 
concentrations to well below those predicted by hydroxyoxide solubility products. 

An exception to that extra-strong attraction may be Fe(III). If constant aeration 
minimizes Fe 2h production, and if biological reactions reach some sort of steady 
state, Fe(IlI) concentrations generally agree closely with those calculated from the 
solubility product of FeOOH. Competing reactions under these conditions appear 
to be overshadowed by the relatively rapid dissolution/precipitation of FeOOH. The 
large amounts of FeOOH in soils, compared to the amount of competing reactions, 
permits the solubility product to control the Fe(III) concentration in soil solutions. 
Aluminium is also prevalent in soils, but the Al(OH)3 solubility product controls 
A\ i+ concentrations in soil solutions only after the cation exchange capacity has 
been saturated with Al 3+ (at uncommonly low pH values). 

Trace metal ions in soil particles are probably contaminants in crystals or in amor- 
phous gels of more abundant ions. The solubilities of such contaminant ions might 
still be definable by solubility products if the cations formed an ideal solid solution 
within the crystal. The solubility should then be a function of their mole fractions 
and solubility products of the minor and major hydroxyoxides. Ideality in such cases 
seems doubtful. 

The solid hydroxyoxides adsorb H + in acid soil solutions and adsorb OH~ (or 
release H h ) in basic soil solutions. The surface charge of such solids changes from 
positive in acid solutions to negative in basic solutions. This behavior is part of the 
reason that soil cation exchange capacities increase with pH, although the contribu- 
tion of hydroxyoxides to soil exchange capacities is small except in highly weathered 
soils. 

The manganese, iron, cobalt, copper, zinc, and molybdenum ions of this group are 
all essential for plants and animals. In addition, vanadium, chromium, nickel, and tin 
are essential for animals. The soil solution concentrations and plant availabilities of 
these ions generally decline with increasing pH. Molybdenum is an exception and 
becomes more available with increasing pH. Reducing conditions dissolve Mn 2+ 
and Fe 2+ . 

Compared to the major fertilizer elements N, P, K, and S, plant deficiencies of 
microelements are infrequent. Plant variety and growth rate seem to be at least as im- 
portant as soil factors in determining microelement deficiencies. Common examples 
are Fe and Zn deficiencies in irrigated crops, especially fruit trees, grown on basic 
soils. Such deficiencies are caused in part by high growth rates and consequently 
high plant demands for these elements. Native plants apparently have adapted to the 
natural rates of trace metal recycling. Plant growth rates without irrigation in arid 
regions are of course considerably slower. 

Deficiencies of zinc, copper, cobalt, and molybdenum have also been recognized 
in some acid soils. The Zn, Cu, and Co deficiencies seem to be attributable to the low 
contents remaining in highly weathered soils. Weathering appears to remove these 
ions faster than Fe, Al, Mn, and other trace metal ions. Zinc and Cu also complex 
with soil organic matter, which does much to control their concentrations in the soil 
solution. Soluble organic anions from decomposing organic matter can dissolve Fe 
and Al and move them downward where they reprecipitate in the soil profile. This 



TOXIC ELEMENTS IN SOILS 55 

is characteristic of Spodosols, or Podzols, and the movement there is restricted to 
10-20 cm downward in the soil profile. 

Attempts to estimate the amounts of microelements available to plants have been 
somewhat successful. Soils to be tested are usually extracted with solutions contain- 
ing chelates, such as DTPA or EDTA. The amount of metal extracted is then corre- 
lated with plant response to fertilization with the element. The solubilities of trace 
metal chelates is much greater than the solubilities of the hydrated (aquated) cations, 
so chelates added to soils can increase the solubility and plant availability of some 
trace metals. The Fe, Cu, and Zn complexes are stronger and form preferentially 
in soils of pH < 7. In basic soils Ca 2 " 1 ' and Mg 2+ , because of their high concen- 
trations, compete with the transition metals for the chelating anions. Some chelates 
form stronger complexes with iron and zinc than EDTA. Hence, they can maintain 
higher Fe, Cu, and Zn solution concentrations in basic soils, where Fe and Zn de- 
ficiencies occur. The chelates are effective until soil microbes degrade the chelates. 
Fertilization with chelates has been limited by their high cost and by their biodegrad- 
ability in soils. 



2.7 TOXIC ELEMENTS IN SOILS 

Although ion retention by soils can cause elemental deficiencies, it also prevents 
excessive or toxic concentrations in most soil solutions. The evolution of life took 
advantage of the naturally low concentrations in water and low plant availability; 
higher concentrations evolved as toxic. Table 2.4 shows the natural soil contents of 
ions that are generally harmless. The upper values are conservative estimates of soil 
contents that might lead to toxicity. All soils contain the "toxic elements," even in 
amounts that are mined as ores, and these concentrations do not necessarily harm 
plants or groundwater. 

This section deals with the elements known as heavy metals. Heavy metal is de- 
fined as any metal having a specific gravity greater than iron (s.g. = 5.5) and is 
often used nowadays to mean toxic metal ions. Toxic organic (carbon-containing) 
compounds, such as pesticides and many industrial and municipal wastes, are much 
more prevalent and present a greater hazard to plants and animals than heavy met- 
als; they are discussed elsewhere. Heavy metals may be toxic when their ions are 
available to plants or are in groundwater at concentrations higher than in native soils. 
Trace metals have been defined as being less than 100 mg kg -1 of soil. 

"Heavy metal-contaminated soils" is a common phrase. Regulatory agencies were 
forced to make criteria for contamination and understandably were very conserva- 
tive. The criterion is basically, "if you eat it and it makes you sick, it is contami- 
nated." By that standard, every soil is contaminated. The real issue is whether soil 
components affect groundwater composition or can be taken up by plants and soil 
fauna. Mixed adequately with soil, metal ions quickly react with the soil and are ad- 
sorbed/precipitated and tend to revert to their native states, and native availability, in 
soils. In the laboratory, Pb added to a soil suspension was adsorbed by a roughly two- 
step reaction. The half-life of the first step of retention was 2 minutes and accounted 



56 



IMPORTANT IONS 



Table 2.4. Natural, and apparently safe, soil and plant concentrations of elements that 
have been designated as being toxic 3 







Total Soil 








Typical 




Plants 




Value 


Range 


Soil Solution 


Range 


Element 


(mgkg -1 ) 


(mgkg -1 ) 


(mgL- 1 ) 


(mgkg -1 ) 


Aluminum 


50000 


10000-200000 


0.1-0.6 


— 


Arsenic 


5 


1-50 


0.1 


— 


Beryllium 


1 


0.2-10 


0.001* 


— 


Cadmium 


0.06 


0.01-7 


0.00 l b 


0.1-0.8 


Chromium 


20 


5-1000 


0.00 l b 


— 


Cobalt 


8 


1^10 


0.0 l h 


0.05-0.5 


Copper 


20 


2-100 


0.03-0.3 


4-15 


Lead 


10 


2-100 


0.00 1* 


0.1-10 


Manganese 


850 


100-4000 


0.1-10 


15-100 


Mercury 


0.05 


0.02-0.2 


0.001 


— 


Nickel 


40 


10-1000 


0.05 fc 


~ 1 


Selenium 


0.5 


0.1-2.0 


0.001-0.01 


— 


Zinc 


50 


10-300 


<0.005 


8-15 



"From W. H. Allaway. 1968. Adv. Agmn. 20:235; R. P. Murrman ajid F. R. Koulz. 1972. Special Report 
No. 171, U.S. Army Cold Regions Research and Engineering Lab, Hanover, NH; and G. R. Bradford 
et al. 1975../. Environ. Qual. 4:123. 

Estimated as 30 times its concentration in seawater. 



for 78% of the Pb removed from the water. The half-life of the second step was 58 
hours. Compared to rates of ion uptake by plants and water movement in soils, both 
rates are fast. In the field, soil retention rates are slower because of less mixing, but 
are fast enough to prevent Pb movement and plant uptake. 

Nonetheless, soil contamination is a concern because mixing with the soil can be 
inadequate. For example, the surface soil next to older wooden buildings can contain 
high Pb because of Pb-containing paint flaking off the walls and because Pb moves 
downward very slowly. Children frequently play next to homes and soil gets in their 
mouths. Industrial waste dumps may contain high concentrations of heavy metals 
and acidic components that increase the metal's solubility in the water leaching from 
the dumps. 

In many cases, soil "contamination" can be overcome by covering the soil with 
an unaffected layer or by thoroughly mixing the contaminated layer with "clean" 
soil beneath. Excavating and moving the material to a landfill does nothing to alter 
its chemical state and can expose many more people than leaving the soil in place. 
"Hazardous waste landfills" can be hazardous because they concentrate the wastes 
and they rely on suspect technology to isolate the wastes. 

Landfill leachate is a potential source of heavy metal contamination and a popular 
horror story in newspapers. Landfills concentrate wastes, the wastes are not always 
mixed thoroughly with soil, and soils have limits in their capacity to react. Landfills 



TOXIC ELEMENTS IN SOILS 



57 



have strong reducing conditions and the reduced oxidation states of metals tend to be 
more water soluble. These conditions, however, also form organic matter and sulfide, 
which react with and inactivate the ions. 

A popular method of minimizing leachate pollution may be counterproductive in 
the long run. Thick liners of polyethylene sheets or clay below the landfill prevent 
water from leaching out, until cracks or leaks occur. Then the water flows down 
through a small soil volume below the leak and may oversaturate the soil. Liners then 
accentuate the pollution hazard. Wise water management and treatment would appear 
to be a better answer than liners. Evidence for landfill contamination of groundwater 
is meager anyway. Liners that encapsulate landfills are an overreaction to a problem 
that is less serious than was perceived. Entombing wastes is an ostrich-in-the-sand 
approach to waste management. Sealing off oxygen input allows only fermentation 
of organic wastes rather than oxidizing them. The goal should be oxidation, that is, 
destruction of waste compounds, and returning their constituent elements to their 
oxidized and most benign states. 

The total amount of a heavy metal or any substance in the soil is a poor indicator 
of its availability to plants or ability to move in soils. Defining the amounts that are 
toxic to plants or to animals subsisting on those plants is very difficult. Plant and soil 
concentrations that are harmful to plants are unknown for most ions and for most 
plants. In addition, the harmful concentrations vary with the species, age, health, and 
general nutrition of the organization. The soil solution column of Table 2.5 gives esti- 
mates of the concentrations that are found in soils and that are immediately available 



Table 2.5. Representative ion concentrations in soil solutions 
of temperate region soils 9 



Ion 


mgL ' 


Molar, xl0~° 


cr 


60-600 


2000-20 000 


S (as SO;") 


50-500 


500-5000 


Ca 2 + 


30-300 


800-8000 


Mg 2+ 


5-50 


200-2000 


Si (as Si(OH) 4 ) 


10-50 


400-2000 


K+ 


1-10 


20-200 


Na+ 


0.5-5 


20-200 


F" 


0.1-0.5 


5-20 


Mo(asHMoO~) 


0.001-0.01 


0.001-0.1 


Mn 2+ 


0.1-10 


2-20 


Cu 2+ 


0.03-0.3 


0.5-5 


P(asH 2 PO-) 


0.002-0.03 


0.006-1 


Al (as A10H 2+ ) 


<0.01 


<0.4 


Fe 2+ + Fe(OH)+ 


<0.005 


<0.01 


Zn 2 + 


<0.005 


<0.01 



"From R. P. Murrman and F. R. Koutz. 1972. Special Report No. 171, U.S. 
Cold Regions Research and Engineering Lab, Hanover, NH. pp. 48-74. 



58 IMPORTANT IONS 



Table 2.6. Maximum recommended concentrations of toxic 
ions in drinking water for livestock 9 

Ion Upper Limit (mg L~ ) 

Aluminum 5 

Arsenic 0.2 

Beryllium No data 

Boron 5.0 

Cadmium 0.05 

Chromium 1 .0 

Cobalt 1.0 

Copper 0.5 

Fluoride 2.0 

Iron No data 

Lead 0. 1 

Manganese No data 

Mercury 0.01 

Molybdenum No data 

Nitrate + nitrite 100 

Selenium 0.05 

Vanadium 0.10 

Zinc 24 

Total dissolved solids 1 000 

"From Water Quality Criteria. Environmental Studies Board, National 
Academy of Science. National Academy of Engineering, 1972. 



to plants. The recommended maxima for livestock drinking water (Table 2.6) are 
very conservative guides to the desirable maxima in soil solutions, because plants 
are much more tolerant of high concentrations than are animals. Plants had to evolve 
a greater tolerance to both toxicity and deficiency because of the limited soil volume 
within reach of their roots. Animals can range over a much wider area. Soil reten- 
tion, exclusion by plant roots, and limited translocation to the plant top all exclude 
soil ions from the animal food web. 

Soil solution concentrations of most trace metals are largely unknown because 
of difficulties in measuring small concentrations. The values in Table 2.4 marked 
with b are only rough estimates derived from the composition of seawater. Reported 
Mn and Cu concentrations in soil solutions are about 30 times greater than their 
concentrations in seawater. This factor was applied to the remaining ions of the table 
as well. 

All of the essential microelements and most, if not all, of the trace elements are 
toxic at soil concentrations much above normal. Naturally occurring high concentra- 
tions of toxic elements are rare in soils, except for widespread Al 3+ phytotoxicity in 
acid soils. Soil contamination by toxic elements generally is a result of human activi- 
ties. Anthropogenic pollutant elements and their important oxidation stales that have 
received attention include, in order of atomic number rather than importance: Be(II), 
F(-I), Cr(lll-VI), Ni(II-III), Zn(II), As(III-V), Cd(II), Hg(O-I-II), and Pb(II-IV). 



TOXIC ELEMENTS IN SOILS 



59 



Soil contamination from smelting, metal plating, manufacturing, municipal and in- 
dustrial wastes, and automobile traffic can increase soil concentrations of these ions 
to possibly toxic levels. Even so, animal problems generally occur only when con- 
taminated plants are the sole food supply. Grazing animals confined to contaminated 
areas show the most serious effects of toxic metal accumulations in soils. 

Figure 2.3 shows the release of six transition metals from ten widely different soils 
of central Europe that had been affected by smelting and other industrial pollution. 
Despite the differences in soils and amounts of exposure, the percent desorbed from 
the soils at constant pH is quite consistent for each ion. In this group Cd is retained 
least strongly and this has been generally true in soils; Ni and Cr are retained the 
strongest in this experiment. The relative positions of the ions change slightly with 
the method of soil extraction. Lead is retained strongly by many soils and chromium 
as Cr(VI) can be more mobile and more available to plants than Fig. 2.3 suggests. 
Ni, Zn, Hg, and Pb availability are generally less responsive to pH than Al and Be. 




10 20 



30 40 50 100 10 20 
Time after starting the tritation (h) 



5 100 



FIGURE 2.3. Average heavy metal release during the pH stat experiment in percentage of the to- 
tal concentration (n = 1 0). Standard deviations are plotted as error bars and show the variability 
between the samples. 



60 IMPORTANT IONS 

The availability of the selenate, vanadate, arsenate, and chromate anions decreases 
with the soil's Fe and Al hydroxyoxide content and increases with pH, since anion 
retention decreases as pH increases. The pH response for anions, however, is gen- 
erally less dramatic than for cations, which adsorb/precipitate as hydroxyoxides in 
soils. 

The metals Hg(O-I-II) and Cd(II) are retained less strongly by soils than the other 
toxic cations, and hence pose a more serious problem of movement and plant avail- 
ability. Cadmium is a rather soluble transition metal that behaves somewhat like 
Ca 2+ except that Cd 2+ reacts more strongly with organic matter and with sulfides. 
By absolute standards, however, Cd movement and plant availability in soils are 
small. Because of its extreme toxicity, Cd 2+ is a serious concern in soils used as 
waste-disposal sites. As might be surmised from the insolubility of Hg(OH) 2 , Hg(II) 
is retained rather strongly by soil. Hg(II) can be reduced to Hg(I), Eh = 0.92 V, and 
to Hg(0), Eh° = 0.79 - 0.059 log(Hg+)/4. The Hg(TI) oxidation state probably pre- 
dominates, but reduction to elemental Hg(0) occurs. Elemental Hg is slightly volatile 
and diffuses as a gas through soil pores. Compared to the other toxic metals in soils, 
Hg is relatively mobile. The toxic compound dimethyl mercury (Hg(CH3)2), formed 
in contaminated and highly reduced aquatic sediments, seems to be rare in soils. 

Time decreases the availability of ions added to soils. Time allows ions to diffuse 
to the strongest sorptivc sites, including incorporation into the surfaces of weathered 
soil particles. Time also leads to the aging of soil solids, with smaller and more 
reactive phases transforming into larger, less reactive, and less plant-available and 
water-soluble particles and organic matter. Leaching of the toxic cations and anions 
(except NO^" and perhaps CrO^) from soils is generally negligible. Toxic elements 
tend to remain within a few centimeters of where they first contact the soil unless the 
soil is stirred by cultivation. If retained at the immediate soil surface, they are above 
the most active portion of the root zone. 

The chromate (CrO 2- ) anion moves through well-aerated soils of moderate to 
high pH. Although chromate is a strong oxidizing agent and hence easily reduced 
at high concentrations, Cr0 4 ~ stability increases at increasing pH and at the dilute 
(I0~ 6 M) Cr concentrations in soil solutions. 

Despite concerns about Cd, Hg, and Cr in soils, the soil is still a far safer medium 
for the disposal of pollutant wastes than any other part of the environment. Soils 
are better able to oxidize (the oxidized state is generally the least toxic state), to 
retain pollutants safely, and to remove them from the food web, than air or water. 
For example, the increase in toxic metal concentration in plants is about one-half to 
one-tenth of the corresponding concentration increase in soils. Assuming a total dry 
matter crop production of 10 000 kg ha -1 and a soil bulk density of 1 .3, adding the 
high amount of 1 30 kg ha - ' of a heavy metal in wastes to the surface meter of soil 
(a mass of 1 300000 kg ha -1 in -1 ) increases the element's concentration in the first 
crop by about 1 mg kg~ ' or 1 g ha~ ' , and that concentration and amount decreases 
with time. The amount in the harvested portion of the plant is much less than that in 
the total plant. 

Municipal waste could be a valuable source of nitrogen, phosphate, and water 
for crops. Under careless management, however, waste application can lead to trace 



CARBON, NITROGEN, AND SULFUR 61 

metal pollution of soils. Part of the answer is to prevent initial toxic metal contami- 
nation of municipal wastes by the relevant industries. Where such contamination is 
unavoidable, the safest procedure might be to dispose the wastes on forested land, 
land dedicated to waste disposal, or where the plants grown are far from a food web, 
and to insure that the wastes are actively mixed and diluted into the soil. 



2.8 CARBON, NITROGEN, AND SULFUR 

These three elements are grouped together because (1) their geochemical cycles are 
rapid and include the atmosphere, (2) they change oxidation states rapidly and cycli- 
cally in soils and the environment, (3) the changes in oxidation state provide the 
energy for life, and (4) C, N, and S processes in soils are closely interrelated. Pho- 
tosynthesis of carbon and its oxidation back to CO2 provides the energy that drives 
life and the N and S reactions. Nitrogen and sulfur oxidation-reduction also involve 
energy changes but the amounts are small in comparison and usually involve a net 
consumption of energy from carbon sources. The changes and turnover rates of C, N, 
and S compounds are sometimes of the order of hours and days. Turnover rate is the 
time required for one complete change of that substance from one phase to another. 
The other elements in the periodic table cycle annually between plants and soil and 
between soils and rocks in geologic time scales. 

In humid soils, the C, N, and S are predominantly in organic compounds, result- 
ing in the C/N/S mass ratio of about 100/10/1. With increasing aridity, the amount of 
soil organic matter decreases and the amounts of carbonate, sulfate, and nitrate an- 
ions in the soil solution tend to increase. In all soils, the N and S cycle through many 
soil/plant/soil cycles before some N2 is formed and lost to the atmosphere. The N2 
later returns to the soil by nitrogen fixation by soil microorganisms, and as nitrate 
formed by lightning and washed out of the air by rain. The exchange of S between 
soil and the atmosphere is much slower. Although H2S may form in soils, it reacts 
rapidly with Fe and other transition metals in soils before being released to the atmo- 
sphere. The fallout of industrially formed sulfate and nitrate from the atmosphere as 
part of acid rain is a modem phenomenon. The nitrate in acid rain is more than that 
formed naturally in the atmosphere by lightning strokes. 

Oxidation slate changes cause great changes in the physical and chemical prop- 
erties of elements. Some of the oxidation states produce volatile compounds (CO2, 
CH 4 , N2O, N2, NH 3 , S0 2 , H 2 0, etc.), so that C, N, and S cycle between soils and 
the atmosphere. Similar exchanges occur between the atmosphere and the oceans, so 
the behavior of C, N, and S is more complex than that of many other elements. Their 
exchanges are also more sensitive to environmental changes. Table 1 .3 shows current 
estimates of the worldwide distribution of C, N, S, and oxygen in the atmosphere, 
the surface hydrosphere (freshwaters and ocean waters < 50 m depth), the biosphere 
(living organisms), and the surface meter of soil. The amount of C as soil carbonate 
is a crude guess; the other estimates are slightly better. The amounts of C and S in 
soil organic matter are almost as large as the sums of C and S in all other reservoirs 
combined. The largest fraction of N, on the other hand, is in the atmosphere. 



62 IMPORTANT IONS 

The annual exchanges of C, N, and S between the active parts of the environment 
(Table 1 .4) are probably more meaningful than the total amounts in Table 1.3 but are 
also more difficult to estimate. The only rates known with any certainty are the rates 
of fossil fuel combustion and fertilizer denitrification. The other rates in Table 1 .4 
may be in error by as much as an order of magnitude. The environmental effects of 
human alterations of the natural carbon, nitrogen, and sulfur cycles are little known. 
Climate changes due to soil emissions of CO2, N2O, and CH4 to the atmosphere 
are being hotly debated. Since water in the atmosphere has a much greater effect on 
atmospheric temperature than the other gases, the soil's effect on plant growth and 
the plant's great effect on water balance may be as important as the gas emissions. 
Soils buffer the composition of the C, N, and S gases in the atmosphere. Soil organic 
matter is the largest carbon reservoir in the environment and its amount changes in- 
versely with the CO2 concentration in the air, increases with rainfall, and decreases 
with increasing temperature. The accumulation of carbon as peat and soil organic 
matter in northern Canada has come at the expense of atmospheric CO2 since the re- 
treat of the glaciers in that area 12 000 years ago. The full extent of soil effectiveness 
in buffering CO2 is poorly understood. 

The oxidation-reduction reactions of C, N, S, and O in soils are, virtually with- 
out exception, catalyzed by microbial enzymes. These reaction rates without such 
catalysis are very slow (irreversible). Even with catalysis, the reactions are quite irre- 
versible. The behavior of carbon and nitrogen in particular is dominated by nonequi- 
librium. 

2.8.1 Nitrogen 

In economic and agricultural terms, nitrogen is the most important element in fertil- 
ization and in optimizing plant yield per unit area. Water availability is more impor- 
tant to plant growth but is outside the scope of this book. Nitrogen in soil has been 
studied for centuries and is still the most-studied element in soil chemistry, micro- 
biology, and fertility. It is the soil element that most commonly limits plant growth. 
Nitrogen reactions are treated in detail in texts on soil fertility and soil microbiology 
and in many review articles. 

Nitrogen chemistry in soils is the changes of nitrogen during organic reactions, 
oxidation of organic nitrogen to N2 and N2O (denitrification) or to NO^ (miner- 
alization), and the reduction of N2 to amino acids (nitrogen fixation). All of these 
steps require microbial catalysis, but are nonetheless irreversible, and N fixation re- 
quires energy considerably in excess of the amount needed for a thermodynamically 
reversible process, that is, microbial N fixation is energy-inefficient. The thermody- 
namically stable states of nitrogen under soil conditions are shown in Chapter 4. Ni- 
trate is stable only under strongly oxidizing conditions, and amino N, under strongly 
reducing conditions. N2 is the dominant species in the environment, but the rates and 
mechanisms of nitrogen reactions inspire much more interest than the stable states. 

Although many microorganisms are apparently capable of denitrification, only 
a few specialized species, including the free-living Azotobacter, blue-green algae, 
some anaerobic bacteria, and Rhizobium bacteria in root nodules of legumes, are ca- 



CARBON, NITROGEN, AND SULFUR 63 

pable of nitrogen fixation. Energy for nitrogen fixation is provided by photosynthesis. 
A goal of plant research is to enlarge the number of plants that can support rhizobial 
nitrogen fixation, so N fertilization would no longer be necessary. If attainable, this 
N fertilization would not be free. Crop yields would decrease by about 25% com- 
pared to N-fertilized plants, because energy would be diverted away from growth to 
reduce N2. 

The dependence of nitrogen transformations on photosynthetic energy gives rise 
to a C/N mole ratio of typically about 12, or mass ratio of 10, in soils. When the 
natural ratio is changed by nitrogen fertilization, plant uptake of nitrogen, or addition 
of low-nitrogen organic matter such as straw and wood, soil microorganisms restore 
the balance by carbon oxidation, nitrogen fixation, or denitrification over the course 
of several weeks to months. Soil microbes resist any long-term changes of the soil's 
natural N status. 

Soil nitrogen contents fluctuate with soil and fertilizer management, including 
growth of leguminous versus nonleguminous crops. The production and use of nitro- 
gen fertilizers is a substantial portion of the nitrogen cycle in agricultural areas. The 
200-Tg yr~ ' rate of terrestrial nitrogen fixation corresponds to about 20 kg ha~ ' y r~ l 
over the earth's land area of 1.2 x 10 8 km 2 . The 30-Tg yr _1 production of nitrogen 
fertilizers is applied to perhaps one-third of the 11% of the land area that is culti- 
vated. The average nitrogen input to the fertilized soils is therefore about 75 kg ha - ! 
N. Roughly half of the fertilizer N is recovered by plants. The remainder is denitri- 
fied or leached from the soil. Fertilization tends to increase the N2O/N2 ratio of the 
gases produced during denitrification. 

Nitrogen is leached from soils as nitrate; NH^ is retained by cation exchange 
but is oxidized eventually to nitrate if not absorbed by plants and microbes. Nitrate 
leaching is a potential pollution hazard to surface waters and groundwaters. Denitri- 
fication and avid plant and microbial uptake of nitrogen tend to minimize the NO^ 
concentrations of soil solutions. Soil solution extracts from fertilized soils contain 
as much as 20 to 40 mg L -1 of NO3 — N (nitrogen in the form of NOjp. The ni- 
trate concentration is often much less below the root zone. The current NO^~ limit 
for drinking water in the United States is 45 mg L _1 NOJ , or 10 mg L _1 NOj — N. 
Higher concentrations can lead to nitrite poisoning (methemoglobinemia or "blue 
baby" syndrome) in infants. 

High NO^ concentrations in groundwaters generally result from the following-. 

1 . Overfertilization with NH^", NO^", and urea fertilizers. In well-aerated soils, 
with adequate moisture and moderate temperature, NH^" and urea are con- 
verted to NO^~ in a matter of weeks. Groundwater pollution from overfertil- 
ization is most pronounced when early-season or over-winter leaching moves 
fertilizer nitrogen below the root zone of young or seasonally inactive plants. 
Some U. S. states have enacted "Best Management Practice" laws for agricul- 
ture, which include provisions to minimize overfertilization. 

2. High organic matter inputs to soils, including waste disposal operations. 
Sewage sludge, manure disposal, and irrigation with wastewaters can load 



64 IMPORTANT IONS 

the soil with more organic nitrogen than can be utilized by plants. When oxi- 
dized, these materials release CO2, N2, and N2O to the atmosphere and nitrate 
to the soil solution. At moderate loading rates, the NO^" can be utilized by 
plants and soil microorganisms, which reduces its concentration to acceptable 
limits before leaching occurs. At higher loading rates, however, careful water 
management is necessary (0 avoid NO3" pollution of water. 
3. Irrigating arid soils with waters containing natural nitrates. Nitrate accumu- 
lation normally is relatively small compared to the accumulation of other salts, 
but may still be sufficient to raise the nitrate concentrations of groundwaters 
and drainage waters to unacceptable levels. If the downward migration of ni- 
trate is slow, the nitrate needs many years before it reaches the groundwater 
table in these regions. That may be sufficient time for some of the nitrate to be 
removed by microbes in the vadose zone (the soil zone between the root zone 
and the groundwater table). 

The complete mechanism of soil nitrogen redox changes is unknown despite much 
study. The overall oxidative reactions are 

Organic-N and NH 3 = N 2 and N 2 (2.5) 

or 

Organic-N and NH 3 = NO" (2.6) 

The amount of N2 and N2O lost during oxidation under natural conditions is usually 
small. These gases are lost in greater amounts when nitrogen fertilizers are added to 
soils, and the fraction lost as N?0 increases greatly. 

The reductive pathways for nitrogen include NOJ and N2 reduction to amino 
acids in plants and in microbes. Soil microorganisms reduce nitrogen stepwise: 

NOJ = NOJ = N 2 or N 2 (2.7) 

or 

NO, = NO^ = organic-N or NH3 (2.8) 

Unknown intermediate steps exist between the nitrite (NO^~), N 2 , and NH3 forms. 
The fraction denitrified and the N2O/N2 ratio increase with increasing amounts of 
initial NO^. Nitrate reduction is the main pathway of denitrification, N2 and NtO 
production, in soils. 

The amounts of NJTJ" and NOJ in soils are small compared to the amounts of or- 
ganic N. Because soil nitrogen contents tend toward steady states, the concentrations 
of NOj" and NH 4 ( ' in the soil solution are rough indicators of nitrogen availability to 
plants. 

Another portion of the nitrogen cycle involves the NH3 that is liberated from soils 
and the oceans and is then reabsorbed. Current estimates are that 1 00 to 250 Tg yr _ ' 
of nitrogen are lost from the world's soils as NH3, that 30 to 50 Tg yr~ ' of NH3 enter 



CARBON, NITROGEN, AND SULFUR 65 

soils in rainwater, and 600 to 1300 Tg yr -1 of NH3 enter the soil-plant system by 
direct absorption of the gas from the atmosphere. Considerable ammonia is released 
from animal feedlots and manure. 

Ammonia concentrations in well-aerated soil solutions are normally <10 -6 M. 
The NHj ion is almost identical in size to K + , and can be held between 2: 1 layer sil- 
icate lattices ("fixed") in the same way as K + . In recently fertilized soils, anaerobic 
soils and sediments, soils used for municipal and agricultural waste disposal, and for- 
est soils of low pH, exchangeable NH 4 h levels can be a significant portion of the soil's 
inorganic nitrogen. Relative to other exchangeable cations, however, the amount of 
exchangeable NHj is small. Ammonium ions are less mobile than NOJ and are less 
likely to be lost through denitrification, although ammonia readily volatilizes from 
the surface of alkaline soils. Nitrification (NH^" to NO^) inhibitors are commercially 
available. They are being investigated as a means of maintaining fertilizer nitrogen 
in the NHj form, thus retarding denitrification and leaching losses. 

2.8.2 Sulfur 

Although sulfur is less volatile than carbon or nitrogen, organic decay in swampy 
soils may release a little sulfur as methyl mercaptan (CH3SH), dimethyl sulfide 
(CH 3 — S— CH 3 ), dimethyl disulfide (CH 3 — S— S— CH 3 ), and hydrogen sulfide 
(H2S), but the release is rare in aerobic soils. The organic sulfur gases are the dis- 
tinctive odor from paper mills. Within the soil, these gases oxidize, or the sulfide 
reacts strongly with transition metal ions in soils and precipitates as FeS2, MnS, 
and other transition metal sulfides rather than escaping to the atmosphere. If soil 
emissions of H2S and the organic sulfide gases were substantial, their strong odors 
would be noticeable. Sulfur emission worldwide from anaerobic soils and swamps 
has been estimated as 30 Tg yr"" ' . 

Sulfate is the stable sulfur oxidation state in aerobic soils, and sulfide is stable 
in anaerobic soils. Sulfur changes its oxidation state by microbial catalysis and the 
changes seem to be much more reversible than nitrogen and carbon reactions. Ele- 
mental sulfur is rare naturally in soils but is sometimes added to soils as an amend- 
ment, and sulfides are common in many mining wastes. When elemental sulfur and 
sulfides are exposed to oxygen, they oxidize to H2SO4. Soil acidities as high as pH 
2 may persist until the sulfide or sulfur has all been oxidized and leached away. 

Major sulfur inputs to soils include atmospheric SO2 and its various oxidation 
products from coal combustion (100 Tg yr -1 sulfur), petroleum processing (30 Tg 
yr -1 ), ore smelting (15 Tg yr -1 ), and sulfate from sea spray (20 Tg yr -1 ). Most 
of the atmosphere's sulfur falls near the areas where it is produced. The sulfur fall- 
out over West Germany, for example, is as high as 50 kg ha"" 1 yr -1 in nonforested 
areas and 80 kg ha - ' yr - ' in forests. The fallout occurs both as acid rain and as di- 
rect plant absorption; direct soil absorption of atmospheric sulfur is minor in humid 
and temperate regions. In arid regions SO2 and its oxidation products, H2SO3 and 
H2SO4, are absorbed directly and rapidly by the basic soils and their dust. The acid- 
ity from the large amounts of S formerly emitted by copper smelters in arid Southern 
Arizona was neutralized in the atmosphere within 50-100 km of the sources. The 



66 IMPORTANT IONS 

largest emitter (2000 tons H2SC>4/day for 70 years) acidified nearby surface soils to 
a depth of 10 mm. 

Atmospheric sulfur from smokestacks can be carried hundreds of kilometers over 
flat terrain. In regions of sulfur-deficient soils, atmospheric sulfur at low concen- 
trations can benefit plants. Benefits from the low concentration, however, must be 
weighed against the associated acidification of freshwater, phytotoxicity, health haz- 
ards, smog, and building deterioration at the higher concentrations near the sources. 

Sulfate anions are retained only weakly by soils, but the retention increases with 
soil acidity. Sulfate anions are absorbed readily by plants and incorporated into 
biomass. Hence, biomass and SOM constitute large sulfur reservoirs at the earth's 
surface. The C/S mass ratio in soil organic matter is typically about 100/1. The sul- 
fate content of soils increases with aridity and with salt accumulation. 

Widespread sulfur deficiencies in many agricultural soils became more obvious 
in recent decades as "treble superphosphate" began to supplant "superphosphate" as 
a phosphate fertilizer. Superphosphate is made with H2SO4 and contains about 50 
mol % sulfate. This inadvertent sulfur fertilization ended when higher purity treble 
superphosphate, made with H3PO4, was substituted. The "improved" treble super- 
phosphate sometimes produced lower crop yields than did superphosphate, until the 
sulfur deficiency was recognized and corrected. 



BIBLIOGRAPHY 

Black, C. A. 1968. Soil-Plant Relationships, 2d ed. Wiley, New York. 

Mortvedt, J. M., P. Giordano, and W. L. Lindsay (eds.). 1972. Micronutrients in Agriculture. 

American Society of Agronomy, Madison, WI. 
Russell, E. W. 1973. Soil Conditions and Plant Growth, 10th ed. Longmans, London. 



QUESTIONS AND PROBLEMS 

1. Why are total soil contents poor indicators of the amounts of ions that may enter 
the food chain? By what mechanisms are ions held by soils? 

2. What are the major exchangeable cations in soils? How do the relative propor- 
tions change between acid and basic soils? What are the rates of exchange for 
the various cations? What soil conditions are typical deficiency, sufficiency, and 
toxicity (if any) for these ions? 

3. Discuss the soil's role in the cycling of a given element in the food chain. 

4. How are Fe(III) and Al(ITT) reactions both different and similar in soils? Which 
transition metals are essential to 

(a) Plants? 

(b) Animals? 



QUESTIONS AND PROBLEMS 67 

Which are only toxic, so far as is presently known? Which can be both essential 
and toxic? 

5. What are the forms of the micronutrients in soil solutions? How does their avail- 
ability change with pH? 

6. For a given chemical element, trace the chemical states through which it might 
pass from an igneous mineral through weathering and back to igneous mineral. 
Discuss both solution and solid states and mention the locations (rock, various 
parts of the soil, marine water vs. sediment, etc.) in which these states exist. 

7. Explain why CaC03 tends to accumulate at some depth in semiarid and arid 
region soils, while more soluble salts tend to be in deeper horizons or absent. 
Why is the CaCOj layer thought to represent the average depth of wetting? 

8. If K + is released both from "fixed" forms and by weathering during the growing 
season, why is exchangeable K + , as determined by a single extraction, well 
correlated with the soil's K-supplying power? 

9. Why is F~ a common constituent of igneous rocks? 

10. Explain in your own words why the "lithosphere" and "soil" values of Table 2.2 
often differ so markedly (different elements may require different answers). 

11. Explain how and why plants tend to be much more tolerant of high trace-metal 
concentrations than are animals. 

12. Why are crop sulfur deficiencies more common now than during the 1950s? 



3 



WATER AND SOLUTIONS 



Soil chemistry centers around the soil solution, the aqueous solution that reacts with, 
and is between, the solid phase and roots and soil microbes. A solution is defined as a 
mixture and usually means an aqueous solution — a mixture of solutes and water. Soil 
and rock minerals can be considered solid solutions, because they contain foreign 
ions that are mixed into the mineral's structure. Soil chemistry involves both aqueous 
and solid solutions. This chapter presents the chemistry of ions reacting with H '", 
OH~, O 2 ^, and H2O in the soil solution and soil minerals. 

The state of ions in aqueous solution is similar to their states in soil minerals and 
solid solutions because both water and soil minerals are dominated by the 2 ~ ion. 
Water is an oxide whose charge is countered by H + . In minerals the 2 ~ charge 
is countered by many other cations. Water reacts somewhat like O 2- with cations, 
but much more weakly. The solute composition of the soil solution is the result of 
competition of the oxide groups in soil solids with H + , OH - , and hbO in the soil 
solution. 

Water molecules also interact with each other. The interaction is indicated by 
water's relatively high boiling point and specific heat (the amount of energy needed 
to raise its temperature). The H2O molecule as written looks similar to HtS, and 
water should boil at a lower temperature because oxygen has a lower atomic weight 
than sulfur. However, H2S boils at a much lower temperature, —61° C, than water 
at 0° C, because the H2S molecules are much more independent and interact very 
weakly. The H + ions in HjS are 180° apart, creating a linear, nonpolar molecule. 
The H + ions in H2O are 105° apart, creating a nonlinear, polar molecule. This is a 
dipole with positive (the hydrogen side) end and a negative (the side opposite the H 
ions) end. Technically, water is a quadropole but that sophistication is unnecessary 
here. The positive end of one water molecule attracts the negative end of another 

68 



WATER AND SOLUTIONS 69 

water molecule. This electrostatic attraction is absent in the nonpolar H2S molecule, 
which has no positive and negative ends. Water molecules cluster together in groups 
averaging about six in number at room temperature. Because of thermal motion, 
the groups continually break apart and reform. In ice the structure is a rigid, open 
hexagonal packing of water molecules. In water, the small groups are thought to 
have a similar structure, but each group is not oriented to the next group. The groups 
can slide closer together so water is denser than ice. Water is like an ice slurry on a 
molecular scale. 

Ions and charged surfaces can break down the "ice slurry" structure of water. The 
electric charges are stronger than dipole forces and tend to pull water molecules away 
from their groups by attracting the positive or negative ends of the water dipoles. 
Solutes and water molecules are constantly in motion, but they remain in the vicinity 
of each other for some period of time. If water molecules remain near an ion longer 
than the time required for the water molecules to dissociate from the water structure, 
the ion will have a sphere of water molecules (a solvation sphere or sheath) around 
itself. The number of water molecules in the closest solvation sphere is called the 
primary hydration number. 

The value of the primary hydration number is in dispute. Different methods of 
measurement yield quite different values because they respond to different strengths 
and times of ion-water interaction. Careful measurements of the hydration num- 
ber of Na + , for example, yielded values of 1, 2, 2.5, 4.5, 6 to 7, 16.9, 44.5, and 
71, depending on the method used. Table 3.1 shows hydration numbers for com- 
mon ions determined by several methods that tend to agree. The last column shows 
Bockris and Reddy's estimates of primary hydration numbers for the univalent 
ions. 

Outside of the primary solvation sphere is a second sphere of water molecules 
also affected by the ion's charge. These water molecules have also been torn away 
from their water structure to some extent, but are not so closely associated with the 
ion. The orientation of water molecules in the primary solvation sphere and the more 
random orientation of water molecules in the secondary sphere essentially dissipate 



Table 3.1. Primary hydration numbers 







From 


Most 




From ion 


thermodynamic 


probable 




compressibility 


calculations 


value 


Li+ 


5-6 


5 


5±1 


Na+ 


6-7 


4 


4±1 


K+ 


6-7 


3 


3 ±2 


F~ 


2 


5 


4±1 


cr 


0-1 


3 


2± 1 


Br" 





2 


2± 1 


r 





1 


1±1 



70 WATER AND SOLUTIONS 

Table 3.2. Crystallographic radii and heats and entropies of ion hydration at 25° C 





Crystallo- 








graphic Ion 


Heat of 


Entropy of 




Radius 


Hydration, AW 


Hydration, A 5 


Ion 


(nm) 


(Umor 1 ) 


(J mol -1 K-') 


H+ 


— 


-1090 


109 


Li+ 


0.060 


-506 


117 


Na+ 


0.095 


-397 


87.4 


K+ 


0.133 


-314 


51.9 


Rb+ 


0.148 


-289 


40.2 


Cs+ 


0.169 


-255 


36.8 


Be 2 + 


0.031 


-2470 


— 


Mg 2+ 


0.065 


-1910 


268 


Ca 2 + 


0.099 


-1580 


209 


Ba 2 + 


0.135 


-1290 


159 


Mn 2+ 


0.080 


-1830 


243 


Fe 2 + 


0.076 


-1910 


272 


Cd 2 + 


0.097 


-1790 


230 


Hg 2 + 


0.110 


-1780 


180 


Pb 2 + 


0.120 


-1460 


155 


Al 3 + 


0.050 


-4640 


464 


Fe 3+ 


0.064 


-4360 


460 


La 3 + 


0.115 


-3260 


368 


F - 


0.136 


-506 


151 


cr 


0.181 


-377 


98.3 


Br~ 


0.195 


-343 


82.8 


r 


0.216 


-297 


59.8 


s 2 - 


0.184 


-1380 


130 



the ion's charge within 1 to 2 nm of the central ion. This charge dissipation allows 
the ions to interact less electrostatically with each other. 

The degree of breakdown of the water structure, plus the degree of formation of 
new ion- or solute-water structures, is measured by the heat of hydration (Table 3.2). 
Heats of hydration are less ambiguous than structural concepts of hydration numbers 
and ion- solute- or clay-water interactions. 

The heat of hydration increases negatively, indicating increased interaction, with 
decreasing ion size for a given valence group, from —506 kJ mol -1 for Li" 1 ", atomic 
number 7, to -255 kJ mol -1 for Cs + , atomic number 133. The heat of hydration 
also increases negatively with increasing ionic charge. The heat of hydration for 
monovalent cations is several hundred kJ mol -1 ; for divalent cations, it is about 
— 1600 kJ mol -1 ; and for trivalent cations, it is about — 4000 kJ mol -1 . 

The highly negative heats of hydration for the trivalent ions support the idea that 
they exist as Fe(H20)^ + and Al(H20)g + (hexaquoiron(III) and hexaquoaluminium) 
ions in water. The ion and water molecules in such complexes are tightly bound and 



ACIDS AND BASES 71 

together behave like a single large ion. Divalent transition metal ions and Mg 2+ form 
similar but weaker structures with water. The other alkaline earth cations are larger 
and associate less strongly with water. 

Soil particle surfaces are also charged so they attract ions and water dipoles. The 
charge can be an atomic layer beneath the crystal surface so the interaction between 
soil particles and water is weaker than between ions and water. At the edges of crys- 
tals, the charge is weaker but at the surface. 



3.1 ACIDS AND BASES 

Substances that liberate H + in solution are Bronsted acids. Strong acids like HC1, 
HNO3, HCIO4, and H2SO4 completely dissociate into H + (which probably exists as 
the hydrated or hydronium ions H30 + or H7O*) and an anion. Weak acids deproto- 
nate less readily. Examples are Al 3+ (p ATi = 5), HF (pA" = 3.2), and organic acids 
such as acetic acid, CH3COOH (pK = 5), where K is the dissociation constant of 
the acids. Their more tightly bound protons dissociate at higher pH than strong acids. 
For the generalized weak acid HA, 

HA = H++A" (3.1) 

and its dissociation, deprotonation, or acidity constant is 

K HA = ^2 (3.2) 

UA HA > 

Just as acids differ in strength, bases similarly differ in their ability to give up hy- 
droxyl ions. The strong bases NaOH and KOH dissociate completely in solution. 
Bases such as Ca(OH)2 and Mg(OH) 2 are only slightly less dissociated in solution, 
but are considerably less soluble than either NaOH or KOH. This lower solubility 
accounts for the lower pH of Ca(OH) 2 and Mg(OH) 2 solutions more accurately than 
does weak dissociation. 

Weak bases are more precisely compounds like NH4OH (pK\, = 5), where p#b 
is the negative log of the equilibrium constant: 

(NH+KOH-) 5 

* NH < 0H = NH4OH = l0 (33) 

The NH4OH molecule is the hydrate of NH3, which is very soluble in water. Most 
of the solute remains in solution as NH3, but a small portion forms NH4OH, which 
dissociates into NH4 and OH - . The denominator of Eq. 3.3 is really the sum of the 
dissolved NH3 plus NH4OH. Dissolved CO2 and H2CO3 are identical with ammonia 
in this respect. 

Because the pH change during titration of acid soils with OH - resembles the 
titration curves of weak acids with OH", soil clays were called weak acids for many 
years. For two main reasons, however, weak acid inadequately describes soil acidity. 



72 WATER AND SOLUTIONS 

Firstly, soil acidity changes slowly with time after liming, for example, while true 
weak acids react instantly with OH - . Freshly prepared acid clays are strongly acidic 
but begin to decompose to weak acids within a few hours. The clay decomposition 
liberates Al 3+ . The base added to the decomposed acid clay actually titrates Al 3+ , 
which has a titration curve resembling that of a weak acid. Aluminium hydrolysis 
and clay decomposition are slow, however, so the suspension pH changes slowly with 
time. Secondly, the amount of soil acidity released depends on the salt concentration 
of the bulk solution, but salt concentration has little effect on true weak acids. The 
term weak acid is better restricted to those compounds whose acidity is independent 
of time and salt concentration, such as acetic acid and phosphoric acid. Soil organic 
matter satisfies the definition of a weak acid much better than do inorganic soil clays. 



3.1.1 Hydrolysis and Deprotonation 

The attraction of cations for water molecules is so strong that the cation's charge 
tends to repel hydrogen ions, or protons, from the water ligands in the solvation 
sphere and makes them Bronsted acids. As the solution is made more alkaline, more 
H + tends to dissociate. An example is phosphoric acid molecule formed by P 5 "'" in 
solution, PO(OH)3, written more familiarly as H3PO4. This molecule is stable in 
acid solutions. The +5 charge has already repelled 5 H ions from the four water 
molecules that originally surrounded P 5+ . The first H + ion from H3PO4 is repelled 
by P 5h at about pH 3 to form HtPO^, the second at pH 7, and the last H+ at pH 10. 

The S 6 + in sulfuric acid (S0 2 (OH) 2 ) and the N 5+ in nitric acid (N0 2 OH) are 
stronger acids, and the H + from these strong acids is repelled even in extremely acid 
solutions. The remnants of the water molecules in PO;, - , SO^ - , and NO^ - are oxide 
(O 2- ) ligands that are almost impossible to strip away from the central ion, unless 
the central ion is first reduced to a lower oxidation state. 

A similar repulsion of H + occurs from the solvation sphere of hexaquoiron(IU) 
ions, 

Fe(H 2 6 ) 3+ = Fe(H 2 0)2+ + H ' (3.4) 

This reaction is called hydrolysis, because it splits a water molecule, or deprotona- 
tion. The hydrolysis constant, ignoring the water ligands, is 

The Fe(H 2 0)^ + ion is a weaker acid than sulfuric or nitric acid, but is a stronger acid 
than phosphoric acid. Table 3.3 gives the hydrolysis constants of several common 
cations. The smaller the hydrolysis constant (the more negative its exponent), the 
weaker the acid. 

The hydrolysis of iron(III) continues progressively at still higher pH, giving 

FeOH(H 2 0)|+ = Fe(0H) 2 (H 2 O)+ + H+ (3.6) 



ACIDS AND BASES 73 

Table 3.3. Solubility products and hydrolysis constants of metal ions* 

log #2 



Ion 


log A'sp 


logffi* 


Bc 2 + 


-21 


-6.5 


Mg 2 + 
Ca 2 + 


-10.8 
-5.0 


-12 
-12.5 


Mn 2+ 

Fe 2 + 
Ni 2 + 


-12.5 
-14.8 

-15 


-10.5 
-7 
-8 


Cu 2 + 
Zn 2 + 
Cd 2+ 


-19.5 

-17 

-14 


-7.5 
-9.1 
-10 


Hg 2+ 
Pb 2 + 


-25.5 
-18 


-3.5 
-8 


Al 3 + 

Fe 3 + 
La 3 '- 


-33.5 

-39 

-20 


-5 

-2.9 

-9 


Ti 4 +((TiO>)(OH) 2 ) 
Th 4+ 


-29 

-44 


> -1 

-4.1 



-5.5 
-3.3 



"From L. G. Sillcn and A. E. Martell. 1974. Stability constants. The Chemical Society Spec. Publ. 25, 
London. 

br _ (MOH ( "~" + )(H l ) 
(M»+) 



'K| 



and 

Fe(OH) 2 (H 2 0)- = Fe(OH) 3 (H 2 0) 3 -f H+ (3.7) 

The uncharged Fe(OH)3(Fi20)3 species does not repel other iron(III)-water ions and 
is therefore less water soluble than the charged ions. It can lose four water molecules 
and precipitate as FeOOH. FeOOH is more stable than Fe(OH)3. These hydrolysis 
reactions of iron and especially of aluminium are primary factors in the production 
and control of soil acidity. 

The hydrolysis in Eqs. 3.4, 3.6, and 3.7 oversimplifies the actual mechanism of 
hydrolysis. The hydroly/ed ions begin to associate and form polymiclear ions, which 
enlarge with further hydrolysis. These polynuclear ions form in solutions and may 
exist in natural waters for a long time before they precipitate as a solid. Whether they 
exist in soil solutions, with the soil's large and adsorptive surface area and at the low 
concentrations in soil solutions, is less certain. When ions dissolve from solids, on 
the other hand, the ions formed most likely remain as mononuclear species because 
their concentrations are so low. 

The hydroxyoxides are capable of still further hydrolysis if the pH becomes more 
alkaline. Iron hydroxide, for example, can dissolve according to the reaction 



74 



WATER AND SOLUTIONS 



Fe(OH) 3 (H 2 0) 3 = Fe(OH) 4 (H 2 Or + H + 



(3.8) 



or 



Fe(OH) 3 (H 2 0) 3 + OH - = Fe(OH) 4 (H 2 0)~ 



(3.9) 



Equations 3.8 and 3.9 are equivalent, because producing 1 mol of H + is the same as 
consuming 1 mol of OH - . The equilibrium constant is 



OH~ 



(3.10) 



The Fe(OH)4 ion, because it is charged, is more soluble than FeOOH. Hence, the 
solubility of Fe(lII) increases at alkalinities above pH 8.5. Figure 3. 1 shows the effect 
of pH on the distribution of aquohydroxyiron(III) species in solution. 

The hydrolysis reactions of Al 3+ are analogous to those of Fe 3+ , but Al 3+ is less 
acidic. The loss of the first H + from A1(H 2 0) 6 + occurs at pH 5. The hydrolysis of 
the second and third protons occurs at slightly higher pH and is complicated by poly- 
merization of the hydrolysis products and slow precipitation of Al(OH)3. Figure 3.1 



-2 - 



-6 



o 

€ -10 



-14 



_ 












- 






Solid phases 

FeOOH 




- 






A8(OH) 3 






- 




\ 4 








- 












- 






^K 




/^ 

^ 
<<«" 


- 














i 


i 


l 1 


1 


1 1 



10 



pH 



FIGURE 3.1. Equilibrium solubility of monomeric Fe(lll) and Al(lll) ions from FeOOH and 
AI(OH)3 as a function of pH. 



ACIDS AND BASES 75 



1.0 i- 




J I L 



J I L 



0001 0.002 0.005 0.01 0.02 005 01 0.2 0.5 

1 

FIGURE 3.2. Ion activity coefficients versus ionic strength /. Calculated from the extended 
Debye-Hueckel equation. 



disregards these complications. The Al(OH)^ begins to form above pH 8, but the 
exact value is uncertain. 

High solubility in acidic and basic solutions, and low solubility at neutral pH, as 
shown in Figures 3.1 and 3.2, is called amphoterism. Amphoteric ions can form pos- 
itive or negative hydroxy complexes depending on solution pH. This allows hydrox- 
yoxides to be both acidic and basic. Al(OH) 3 consumes H + when dissolving in acid 
solutions, thereby acting like a base. By consuming OH~ and forming Al(OH)^, alu- 
minium hydroxide acts like an acid. This amphoteric behavior is important to charge 
development by some soil clays. 

The hydrolysis illustrated for Fe 3+ and Al 3+ can happen to all cations, but the 
extent of hydrolysis varies widely. Hydrated alkali and alkaline earth cations should 
deprotonate at extremely high pH, but sufficiently high pH values do not occur in na- 
ture. In typical soil solutions of pH 5 to 9, P(V) exists as H2POJ and HPO^ in aque- 
ous solutions, Fe(III) exists as FetpH.)^ , and Ti(IV) probably exists as TiCKOH)^ - • 
Si(IV) in aqueous solution is Si(OH) 4 , also written FUSiC^; it loses its first H + only 
above pH 9. The pH values at which H + ions are lost from the solvation sheath are 
indicated by the negative logarithms of the hydrolysis constant, p/C, and are shown 
in Table 3.3; pATji < 7 indicates that the cations are acidic. 



3.1.2 Solubility Products 

Table 3.3 gives the solubility products of the metal hydroxyoxides that are likely to 
exist in soils. The general form of the solubility products is 



Ks P = (M" + )(OH-) 



(3.11) 



76 WATER AND SOLUTIONS 

A solid also becomes less water soluble (the solubility product decreases) when its 
crystals are purer, their structure is more ordered, their size increases, and as the 
crystals contain less water (less hydrated), but these effects are secondary to the sol- 
ubility product. The solubility product decreases as crystals grow in size and lose 
waters of hydration and occluded or coprecipitated ions. The slow growth and re- 
crystallization is much more pronounced in the mixture of ions in soil solutions than 
in the pure aqueous solutions of the chemistry laboratory. The solid-phase reactions 
are often exceedingly slow in soils compared to the formation rates of new, poorly 
crystalline material. Hence, soil-formed crystals tend to be small and amorphous and 
to contain many impurities. 

Solubility products measured in pure systems may not represent soil conditions 
very well. The impurities in solids affect their aqueous solubility; soil minerals are 
characteristically impure. Nonetheless, predictions of soil solution concentrations 
usually assume the solubility products of pure minerals apply. The water molecule is 
ignored in stability constant and solubility product equations. The concentration of 
water is assumed to be unity because water is present in great excess and does not 
change significantly during the reaction. This assumption is good in all but the most 
concentrated aqueous solutions and in dry soils. 

3.2 CHEMICAL ACTIVITY 

Ions in water are not free and unattached. They interact with water and with each 
other. Close-range (<0.5 nm) ion-ion interactions are termed complex ions or ion 
pairs and are governed by specific interactions between ions. These close-range in- 
teractions are discussed later. Longer-range (>0.5 nm) interactions are treated by the 
concept of chemical activity. 

The interactions of ions with water molecules and other ions affect the 
concentration-dependent (colligative) properties of solutions. Colligative proper- 
ties include osmotic pressure, boiling point elevation, freezing point depression, 
and the chemical potential, or activity, of the water and the ions. The activity is 
the driving force of reactions. Colligative properties and activities of solutions vary 
nonlinearly with concentration in the real world of nonideal solutions. 

Solutes can be thought of as ideal by considering their activities rather than their 
concentrations. The activity is defined as 

a = yM (3.12) 

where a is the activity, y is the activity coefficient, and M is molarity. The activity 
can be regarded as an ideal concentration, but is more correctly defined as a ratio 
related to concentration and is unitless. The units of y are then L mol -1 . 

The colligative properties of an ideal solution are equal to the concentrations of 
the components, and their activity coefficients equal one. The deviation of the activity 
coefficient from one expresses the degree of nonideality. Figure 3.2 shows the change 
in the aqueous activity coefficients of several ions over the concentrations found in 
soil solutions and groundwater. 



CHEMICAL ACTIVITY 77 

The activity coefficient is defined so that 

lim y = 1 (3.13) 

As the concentration approaches zero, y approaches one. As the solute is diluted by 
the solvent and the solute ions or molecules are farther apart, they interact less with 
each other and behave more ideally. 

Water and ions are affected by the amounts and charges of all of the ions in the 
solution; the ionic strength I combines the effects of concentration and ion charge 

7 = ^E MiZ i 2 (3 - 14) 

where M\ is the molarity, and Z\ is the charge, of each ion i. The ionic strength 
estimates the effective ion concentration by taking into account the large effect of 
ion charge on solution properties. A solution has only one ionic strength, but each 
ion may have a different activity coefficient (Fig. 3.2). 

The properties of individual ions cannot be measured; we can only accurately 
measure the properties of salts and from them estimate the properties of ions. From 
a fundamental or thermodynamic standpoint, ions, therefore, do not exist, but we 
accept that they exist. We estimate the activities of ions by arbitrarily dividing up the 
activities of their salts. The activity of CaCh in a CaCl2 solution, for example, is 

fl&iClo = yea • M C3 2+ (ya ■ M cr ) 2 (3.1 5) 

remembering that the molarity of CP is twice that of Ca 2+ . The interaction between 
ions increases with concentration and with the square of the ion charge. 

The activity of a salt can then be divided between cation and anion. By measuring 
the activities ol'NaCl, NaBr, NaNC>3, and NaClCXi, for example, the contribution of 
Na + to the total activity can be sorted out. This process is very laborious; calculating 
the ion activity directly would be much quicker. 

Debyc and Hueckel in 1924 proposed Eq. 3.16 using the ionic strength to account 
for the effect of all the ions in solution on the activity coefficient of ion i: 

\ogy = -AZff 1 ' 2 (3.16) 

where y is the ion's activity coefficient, A is a constant (= 0.51 1 tor aqueous solu- 
tions at 25° C and is relatively insensitive to temperature), Z\ is the ion's charge, and 
/ is the ionic strength. Equation 3.16 is called the Debye-Hueckel limiting law, be- 
cause it predicts ion activity coefficients only in dilute solutions (solutions that have 
been cynically called "slightly contaminated distilled water"). Equation 3.16 was 
nonetheless a great breakthrough in the understanding of ion behavior in solutions. 

Debye and Hueckel assumed that ions interact electrostatically like charged par- 
ticles of zero size, but ions and their associated water molecules have significant 
physical size. These and other problems make Eq. 3.16 unreliable at ionic strengths 
greater than about 0.0 1 . The equation has been modified to be useful up to / = 0. 1 : 



78 WATER AND SOLUTIONS 



r2 



/1/2 



l ^ = - AZ '\T^rm) (3I7) 

where B = 0.33 for aqueous solutions at 25° C and a\ is an individual ion parameter 
determined experimentally. Table 3.4 gives a\ values for some common ions. Because 
values of a\ somewhat resemble the diameters of ions plus their associated water 
molecules, a\ is thought by some to have physical significance and has been termed 
the "distance of closest approach." 
The empirical Davies equation 



72 



/1/2 



logy ^AZfi-j^-j^- 0.31 (3.18) 

also yields satisfactory values for individual ion activity coefficients over the range 
of concentrations normally encountered in soil solutions and freshwaters. 

Debye and Hueckel's work is historically important to soil chemistry because 
their derivation was similar to that of Guoy and Chapman, published independently 
about 1910, who tried to predict the ion distribution in the aqueous so)ution around a 
charged surface such as a soil particle. Although the Guoy-Chapman theory and its 



Table 3.4. Values of a„ the Debye-Hueckel "distance of closest 
approach" 3 





«/ 


Inorganic Ions 


(I(r 9 m) 


NH+ 


0.25 


cr,NO-,K + 


0.3 


F~,HS-\ OH- 


0.35 


HCO^~, H 2 PO~, Na+ 


0.4-0.45 


HPO^,PO^,S02- 


0.4 


coj~ 

Cd 2+ , Hg 2 +, S 2 - 


0.45 


0.5 


Li+, Ca 2 +, Cu 2+ , Fe 2+ , Mn 2+ , Zn 2+ 


0.6 


Be 2+ , Mg 2+ 


0.8 


H+, AJ 3 +, Fe 3 +, La 3 + 


0.9 


Th 4 +, Zr 4 + 


1.1 


Organic ions 




HCOO- 


0.35 


CH3COO-, (Coor- 


0.45 


dinate^ 


0.5 


C 6 H s COO- 


0.6 



"From i. Kielland. 1937. J. Am. Client. Sot: 59:1675-1678. 



COMPLEX IONS AND ION PAIRS 79 

later modification have shortcomings for soil-water systems, Debye-Hueckel suc- 
cessfully describes the less complicated conditions of dilute aqueous solutions. 

Nonelectrolytes — dissolved gases, organic molecules, neutral ion pairs, and 
undissociated weak acids and bases — are also nonideal solutes in water and are 
common constituents of soil solutions. Their activities also vary nonlinearly with 
concentration, particularly at high concentrations. The activity coefficients of non- 
electrolytes at low concentrations are approximated by 

logy = -W (3.19) 

where k m is called the salting coefficient and / is ionic strength. Measured values of 
k m range from 0.01 to 0.2 foT common nonelectrolytes. The name salting coefficient 
comes from the tendency of nonelectrolytes to be less water soluble at increasing salt 
concentration, so that nonelectrolytes can be "salted out" of solution. 

Like ions in aqueous solutions, the chemical activities of ions solids, and of wa- 
ter, also vary with concentration. Over the range of solute concentrations in soil 
solutions, however, the activity of water changes only negligibly from that of pure 
water. The chemical potentials of pure solids are defined as one, because any amount 
of the solid fixes the equilibrium activity of that substance in the aqueous solution. 
The activity of the aqueous solution is therefore independent of the amount of solid 
present. 

The chemical potential of solid solutions, including impure minerals such as those 
in soils and rocks, is more difficult to define. Isomorphously substituted ions in a 
mineral change its activity and aqueous solubility from that of the pure mineral. 
Progress in defining solid activities has been slow. Soil minerals have often been 
assumed, by necessity, to be pure minerals and assumed to have activity = I. This 
assumption is weak and is discussed later in this chapter. 



3.3 COMPLEX IONS AND ION PAIRS 

When ions and molecules interact closely, they lose their separate identities and are 
better thought of as complex ions or ion pairs. The Al(H20)j? + and Fe(H20)jj + ions 
are complex ions: The water molecules are closely attached to the central ion, with 
the group acting as one entity. 

Complex ions are the combination of a central cation with one or more ligands. 
A ligand is any ion or molecule in the coordination sphere around the central ion, 
H2O in the case of FeOF^O)^' 1- . Water is usually taken for granted in complex ions 
and often disregarded. Ligands replace one or more of the water molecules in the 
primary hydration sphere. Ion pairs, on the other hand, are thought to form by ligand 
attachment in the second solvation sphere (Figure 3.3) and the bonding is weaker 
than in complex ions. Complex ions and ion pairs are synonymous with inner- and 
outer-sphere complexes. Many alkaline earth and transition metal cations are present 
in soil solutions as complex ions or ion pairs. 



80 WATER AND SOLUTIONS 




©(J), 



ION PAIR COMPLEX ION 

FIGURE 3.3. Diagram of an ion pair and a complex ion. 



To associate with a central ion, ligands must compete with the water molecules in 
the central ion's solvation sphere and must lose some of the water molecules in their 
own solvation sphere. In addition, since many ligands are the anions of weak acids, 
H + competes with the central cation for the ligand. Forming a complex ion or ion 
pair involves competition between the cation and H + for the ligands, and between 
the water, OH - , and ligands for the central cation. 

The strength of association between the ions in solution is expressed by various 
equilibrium constants. Stability (formation) constants refer to complex ions and ion 
pairs; hydrolysis (deprotonation) constants refer to the loss of H + from the water lig- 
ands surrounding central cations. Solubility products refer to the aqueous ion activi- 
ties in equilibrium with solid phases. Some "constants" are reported in the literature 
in terms of concentrations rather than activities. Such constants are misnamed, since 
they depend both on the concentration and on the nature of other ions in solution. 
Converting concentrations to activities gives a much more useful value. 

The formation of complex ions is the result of cation-anion attractive forces win- 
ning out in the competition between cations and H + for the various ligands, including 
water. An example is the formation of the monofluoroaluminium complex ion 

A1(H 2 0) 31 ' + F" = AIF(H 2 0)f* + H 2 (3.20) 

This reaction is exploited to extract reactive Al 3+ from soils. Forming A1F 2+ lowers 
the activity of Al(H20)j? + in the water so the fluoride dissolves some Al 3+ from the 
solid phase. The stability constant of this complex ion is 

*aif 2+ = ~4r ( 3 - 21 > 

Air (A1 3+ )(F-) 

The waters of the hydration sphere are usually ignored in the equilibrium constant, 
because excess water is present in aqueous solutions, and the energy of the Al-F 
bond must be greater than that of the ion-water bonds for the complex to form. The 
concentration of the A1F 2+ complex ion increases with increasing concentrations of 
Al 3+ andF~. 

At the same time, H + competes for the fluoride ion. HF is a weak acid and its 
dissociation or acidity constant is 

* HF= <S^P ,3.22, 

HF 



COMPLEX IONS AND ION PAIRS 



81 



Substituting Eq. 3.22 into Eq. 3.21 and rearranging yields 

(A1 3+ )(HF) 



(A1F 2+ ) = K AW 2 + K (W 



H+ 



(3.23) 



The concentration of the A1F 2+ complex ion increases with increasing HF and Al 3+ 
concentrations and decreases with increasing acidity. 

Increasing F _ concentrations encourage more F _ ligands to replace water ligands 
around Al 3+ , to a limit of A1F^~. The hexafluoroaluminium ion is, in fact, the com- 
plex ion removed during fluoride extraction of aluminium from soils, because of the 
high fluoride concentrations employed. 

Ligands such as H2O, OH - , F~, and CN~ occupy only one position around a 
central cation (Fig. 3.4) and are called unidentate ligands. Four of the six F~ ligands 
are in the plane of the Al 3+ cation; the other two ligands are above and below the 
plane. Bridging ligands such as O , CO3 , and PO;, can occupy one position in 
the coordination spheres of two different cations. This produces a polynuclear com- 
plex (nuclear referring to the central ion). The solubility of polynuclear complexes 
is usually less than mononuclear complex ions. Polynuclear complexes tend to poly- 
merize further and precipitate from solution. 

A polydentate ligand can occupy two or more positions around a cation and can 
surround the cation. Such ligands are usually large organic molecules called chelates, 
from the Greek word for claw. Some enzymes, for example, are polydentate ligands 
occupying several positions around a central cation while also bonding to substrate 
molecules. This results in a configuration that catalyzes chemical changes in the sub- 
strate. Soil organic matter strongly adsorbs Cu 2+ , Zn 2+ , Fe 2+ , and other transition 
metal ions, probably by acting as a chelate. 



F" — 



AJ 3 * 



uniden'ole ligands 



H,0 



H 2 — 



HjO- 



Ai 3 * 



-poj- 



I 
HjO 



H 2 



H,0 



AJ 



— H 2 

» \ 



H 2 0- 



I 

HjO 



bidentote (P0 4 5_ ) ligand 



-H,0 



H 2 h 2 



N 

/I 

HjC CH 2 

/ I 

C £ ■■■ 
00 



. Fe< 



N 

l\ 

CH 2 CH 2 



cheloting (EDTA) ligond 
FIGURE 3.4. Schematic structure of the AIF|", binuclear AJ-PO4, and EDTA-iron(ll) complex 



ions. 



82 WATER AND SOLUTIONS 

An example of a chelating ligand is ethylenediaminetetraacetic acid (H4EDTA) 
and its many relatives (Fig. 3.4). The six positions around Fe 2+ are occupied by the 
two amine and four acetate groups. Chlorophyll and hemoglobin are also chelates. 
Chelates are quite soluble and tend to keep Fe, Zn, and Cu in solution for plant 
absorption. Chelates are also used to extract microelement and heavy metal ions 
from soils. The stability constant of the Fe(III)-EDTA complex is 

FeEDTA - 

A-FcEDTA = ,„ •»■.. ^^.a — (3.24) 

(Fe 3+ )(EDTA 4_ ) 
The competing reaction for Fe(III) in soils is its dissolution/precipitation as FeOOH: 

Fe 3+ + 3H 2 = Fe(OH) 3 + 3H + (3.25) 

The solubility product of Fe(OH) 3 is 

K, p = (Fe 3+ )(OH") 3 = (Fe 3+ ) f^) (3.26) 

where K sp is the solubility product of Fe(OH) 3 and K w is the dissociation constant 
of water, tf w = (H + )(OH _ ) = 1(T 14 . 

H + competes by reacting/dissociating as H4EDTA: 

H4EDTA = 4H+ + EDTA 4- (3.27) 

where the dissociation constant is 

(H+)4(EDTA 4 -) 
^ EDTA = H 4 EDTA (3 ' 28) 

Substituting Eqs. 3.26 and 3.28 into 3.24 and rearranging gives the solubility of the 
Fe(IlI)-EDTA complex ion in equilibrium with Fe(OH) 3 : 

(FeEDTA) = K heEDTA ^p # H ,EDTA YU ^ £A . H+ (3.29) 

where K w is the dissociation constant of water. Equation 3.29 can be further extended 
to include the reduction of Fe(III) to Fe(ll). This reduction changes the solubility of 
hydrated iron and hence the stability of the EDTA complex. 

The relations between EDTA and other chelates with cations in pure solution 
can often be applied to soil solutions by including suitable solid-phase controls and 
competition from other cations and hence related to ion uptake by plants. Lindsay 
and co-workers (1979) have carried out such calculations in detail. 

At equal cation concentrations, transition metal cations compete more effectively 
for ligands than can alkali and alkaline earth cations. Transition metai ions have 
the advantage of being able to shift some electrons to better accommodate ligand 
configurations. The ability of unidentate ligands to shift electron orbitals and thus 
form stronger complex ions generally increases in the following order: 1" < Br - < 



COMPLEX IONS AND ION PAIRS 83 



Table 3.5. Stability constants of EDTA (ethylenediamine 
tetraacetic acid) 3 

Metal Ion log K 

LP Z8 

Na+ 1.7 

Mg 2+ 9.0 

Ca 2 + 10.7 

Ba 2+ 7.8 

Mn 2 + 13.8 

Fe 2+ 14.0 

Co 2 + 16.0 

Cu 2 + 18.5 

Zn 2+ 16.3 

Al 3+ 16.1 

Fe 3+ 25.00 

La 3+ 15.4 

Th 4+ 23.2 

( P /C| = 2.0, pK 2 = 2.7, pK 3 = 6.2, p£ 4 = 10.3) 

"From L. G. Sillcn and A. E. Martell. 1974. The Chemical Society Spec. 
Publ. 25, London. 



Cl _ < F~ < C 2 H 5 OH < H 2 < NHj < ethylenediamine < CN~. The strength of 
nitrogen-containing ligands is noteworthy in this list. 

The relative ability of the transition metal ions to form complex ions is Mn 2+ < 
Fe 2+ < Co 2+ < Ni 2+ < Cu 2+ > Zn 2+ for the divalent cations and Cr 3+ = 
Mn 3+ > Fe 3+ < Co 3+ for the trivalent cations. The strongest complexing divalent 
cation is Cu(II). Fe(III) is the weakest complexing trivalent transition metal ion, but 
is stronger than other trivalent cations such as Al 3+ and the lanthanides. The heats 
of hydration (Table 3.2), strengths of EDTA complexes (Table 3.5), and solubility 
products of metal hydroxyoxides (Table 3-3) also follow this general order, with 
water, EDTA, and OH - as the respective ligands. Stability constants less than 10 y 
indicate the weaker ion-ion interaction of ion pairs. 

Figure 3.5 shows how the distribution of EDTA complex ions changes with pH 
under representative soil solution conditions. The changes are doe to competition be- 
tween the cations for all of the ligands. The Fe-EDTA complex predominates in acid 
solutions, because of the great stability of Fe-EDTA complexes and weak compe- 
tition from the low OH" concentrations for Fe(lII). The EDTA ligand prefers Fe(II 
and III) despite the high Ca 2+ and Mg 2+ concentrations in soil solution. Other tran- 
sition metal ions are generally in low concentrations in soil solutions. In alkaline 
soils, however, the higher Ca 2+ and Mg 2+ concentrations and the very low solubil- 
ity of Fe(III) hydroxide favor the formation of CaEDTA and Mg-EDTA complexes 



84 



WATER AND SOLUTIONS 



10° 



<< 

I- - 

o 



lO-'r 



5 10 



i-2 



10' 





i r 


— i 1 1 


i ■ i 
\f Co 


— i — 


; 




Fe 


- 








- 


■ 








Mg 


r 










. 










- 








- 


- 








- 


- 








- 


- 








- 


- 








Ss v s ; 


- 


Al 


1 1 


./ A . i 


■ 



6 7 
pH 



FIGURE 3.5. Mole fraction of EDTA in various complexes versus pH of hypothetical soil solu- 
tions. (From W. A. Norvell. 1974. In Micronutrients in Agricultures J. J. Mortvedt, P. M. Giordano, 
and W. A. Lindsay, (Eds.) Soil Science Society of America, Madison, Wl.) 



instead. This general picture holds true for many compiexing ligands but shifts ac- 
cording to the values of the specific stability constants. 

Another illustration of competition is during laboratory measurements of Ca 2+ 
and Mg 2+ in soil extracts. To ensure that all Ca 2+ , Mg 2+ , and EDTA, but only these 
ions, are present as complex ions, the solution is made alkaline to about pH 10 to pre- 
cipitate transition metal ions. Then CN _ (which complexes strongly with the tran- 
sition metals but weakly with Ca 2+ or Mg 2+ ) is added to complex any remaining 
transition metal cations. 

Ion pairs, or outer-sphere complexes, are written as CaSO^ and CaCO° to dis- 
tinguish them from their respective solids. The CaSO^ and CaCO° ion pairs have 
been found to be particularly important in accounting for apparent supersaturation of 
CaCC>3 in groundwaters and drainage waters of arid regions. Magnesium also forms 
sulfate and carbonate ion pairs in many natural waters. Ion pair formation increases 
with increasing ion charge and concentration. Alkali metal ions such as Na + and K + 
form ion pairs only in highly saline soils and brines. Taking ion pair and complex 
ion formation into account has greatly increased our knowledge of the solid phases 
that govern ion concentrations in soil solutions. 



3.4 HARD AND SOFT LEWIS ACIDS AND BASES 

The reactions of water, H + , OH~, and O 2- describe the aqueous solution behavior 
of many cations — alkali, alkaline earth, Al, and others. This is Bnmsted acid-bane 



SOIL REACTION COEFFICIENTS 85 



Table 3.6. Hard and soft Lewis acids and bases 



Lewis Acids Lewis Bases 

Hard 

H+, Li+, Na+, K+ H 2 0, OH", O 2 ", CO 2- , PO 3- , SO 2 ;" 

Mg 2+ , Ca 24 ", Sr 2 + SiO^~ , F~ , NH3 

Al 3+ , Be 3 " 1 ", Si 4 +, Ti 3 ~ 4+ aluminosilicates 
Mn 2 +,Fe 3+ , C^+.O 3 " 1 - 

Intermediate 
Fe 2+ , Co 2+ , Ni 2 +, Cu 2+ NO", SO^~, CI"", Br"", pyridine 

Zn 2 +, Pb 2+ soil organic matter 

Soft 
Cd 2+ ,Hg l ~ 2+ ,Cu + ,Ag+ S 2 -,CN- 



behavior. These cations also react strongly with oxygen-dominated ligands such as 
CO?, - , SO^ - , NO J, and silicates. These cations are the predominant exchangeable 
ions and the soluble and exchangeable anions in soil solutions. 

The above cations and anions interact weakly, however, with another interesting 
ion group — Cu 2+ , Cd 2+ , Hg 2 " 1 ', S 2 ~, CN~, and other organic groups — which tends 
to react within itself in preference to the oxygen-dominated group. Pearson called 
the oxygen-dominated group hard Lewis acids (cations) and bases (anions) and the 
second group, soft Lewis acids and bases (Table 3.6). Pearson suggested the general 
rule: "Hard Lewis acids tend to associate with hard Lewis bases; soft Lewis acids 
tend to associate with soft Lewis bases." 

Hard Lewis acids and bases have inflexible electron orbitals that form ionic bonds. 
The electron orbitals of soft Lewis acids and bases are more polarizable and more 
likely to form covalent bonds. Soft Lewis acids and bases are also called covale-nt- 
bonding ions and are "siderophile" (sulfur-loving) ions in the geology literature. Or- 
ganic ligands and soil organic matter range from hard to soft Lewis bases. 

This classification explains why, for example, Fe 3+ reacts differently than Fe 2+ in 
soils. Reduced oxidation states tend to be softer Lewis acids and bases. Hard and soft 
also explains why Cd 2+ reacts quite differently than other cations of similar charge 
and size such as Ca, and why soil organic matter reacts with soft Lewis acids and 
also contributes greatly to the exchange capacity of hard Lewis acids. 



3.5 SOIL REACTION COEFFICIENTS 

Many of the hydroxyoxides listed in Table 3.4 exist in soils. Their ion activity 
products, as well as those of phosphates, carbonates, sulfides, and silicates, have 
been measured in soil solutions. Unfortunately, the ion activity products often differ 
widely from accepted solubility products in pure solutions, and also from soil to 
soil. The differences between ion activity products and solubility products is due 



86 WATER AND SOLUTIONS 

to nonequilibrium and the formation of solid solutions whose aqueous solubilities 
differ from the solubilities of pure compounds. The lack of equilibrium is due to 
slow diffusion of ions in the weathered surfaces of soil particles. As the ion diffuses 
into the denser, more crystalline and less weathered interior, its diffusion rate slows 
dramatically. Diffusion rates in the truly solid phase are essentially zero at room tem- 
perature. Diffusion is only significant where water enters into the solid and where 
weathering breaks up the crystal structure. 

The ion activity products in soil solutions and tabulated solubility products should 
agree only if the solubility of a single phase dominates the system, if competing reac- 
tions are insignificant, if the single phase is reasonably pure, and if the system is close 
to equilibrium. This apparently holds true for gypsum, CaS0 4 • 2H2O, and FeOOH. 
Measurements of the (Fe)(OH) 3 ion product in soil suspensions agree fairly well 
with the solubility product of amorphous FeOOH, K sp = 10 -39 . Apparently Fe(III) 
reacts rapidly and only with OH - so that a single solid controls its solubility, includ- 
ing ion-exchange equilibria. Aluminium concentrations, on the other hand, appear 
to be controlled by slow processes, including weathering of aluminosilicates. The 
solubility product of gypsum probably holds because its solubility is great enough to 
swamp out any competing reaction. 

Phosphate is associated with many phases of the soil, including organic matter. 
None of these phases predominates in all soils, and all have different dissociation 
strengths for phosphate. Hence, each should support a different phosphate concen- 
tration, and the strength of association decreases as the phosphate concentration in- 
creases for all of the phases. As a result, phosphate ions should distribute themselves 
among the various retention sites until, at equilibrium, all the ions have the same 
dissociation energy. The speed of these transformations may control soil phosphate 
concentrations rather than the equilibrium solubility of this distribution. 

The rates of soil phosphate reactions also may differ from the rates of phosphate 
uptake by plants and of phosphate release by organic matter decay. This phosphate 
turnover would further upset soil phosphate equilibria. If a steady state (concentra- 
tion is constant with time) existed between the soil and dissolved phosphate ions, it 
might be described by a reaction such as 

soil-OH + H2PO4 = soil-H 2 P0 4 + OH" (3.30) 

where all of the many phosphate interactions with soil are combined into a single 
generalized reaction, so that 

= (soil-H 2 PQ 4 )(OH- ) 
(soil-OH) (H 2 PO") 

where K, is a reaction coefficient rather than an equilibrium constant. Equations 3.30 
and 3.31 can be misleading. They have the form of equilibrium equations, but equi- 
librium does not exist. Also, the activities of soil-adsorbed ion such as soil-OH and 
soil-H2P04 cannot be defined precisely. Equations like 3.30 describe an ongoing 
process rather than equilibrium. 



THERMODYNAMICS 87 

3.6 MODELS OF THE SOIL SOLUTION 

Strictly speaking, soils are always nonequilibrium systems. With care, however, a 
partial equilibrium or steady state can be attained by assuming that the soil solids do 
not change. This is the usual assumption in cation exchange and adsorption studies. 
Kittrick and co-workers were able to obtain near-equilibrium measurements of some 
soil minerals in studies requiring several years. From the resulting ion activities in 
solution, they were able to calculate some of the equilibrium constants used for the 
mineral stability diagrams shown later in this book. 

Ion hydrolysis and solid dissolution reactions occur at the same time in the soil 
solution and many of these reactions are interdependent. One hydrolysis reaction 
that releases H + , for example, affects the other hydrolysis reactions and solids con- 
taining OH - ligands. The advent of computers allowed rapid calculation of many 
simultaneous reactions, and this was soon applied to models that try to calculate the 
composition of the soil solution and natural waters. 

The early models yielded approximate concentrations that reflected the under- 
standing of the soil solution at the time. Later models have yielded better predictions 
of the soil solution's composition, but they are still only approximate. That reflects 
the complexity of the soil more than the inadequacy of modeling. The models pre- 
dict ion interactions in the aqueous solution quite well. Reactions at the surface of 
colloidal particles are more complex, less understood, slower, and hence are more 
difficult to formulate. In addition, the models are forced to use the solubility prod- 
ucts of pure, simple solids. Soil inorganic particles are far from pure compounds, are 
often poorly crystalline to amorphous, are not at internal equilibrium, and may not 
be in equilibrium with the aqueous phase. In addition, the reactions of soil organic 
matter are not known quantitatively; and soils are open systems, meaning that matter 
is continually being added and removed. 

The models therefore reflect our level of knowledge of the soil solution and its 
interaction with soil solids. Since these models have the potential to predict the com- 
position of natural waters (groundwater, lakes and streams, oceans as well as the soil 
solution), soil fertility, the effects of fertilizers and soil amendments, the effects of 
acid rain, and the attenuation and release of pollutants in soils, this important area 
of research should be actively pursued. The accuracy of the models, however, is still 
based on our understanding of the soil's chemistry and cannot be more accurate than 
that. 



APPENDIX 3.1 THERMODYNAMICS 

Chemical activity, heat of hydration, and equilibrium constant are parts of the very 
useful discipline called thermodynamics. Thermodynamics is the relation of matter 
and energy that predicts the direction and final result of chemical and physical reac- 
tions, but does not predict the rate or the path of reactions. This section introduces 
thermodynamic terms commonly encountered in the soils literature. Thermodynamic 
relationships are derived in detail in many physical chemistry texts. 



88 WATER AND SOLUTIONS 

Three disarmingly simple "laws" of thermodynamics that describe the relations 
of matter and energy have evolved into an elaborate structure that can obscure their 
initial simplicity. If an exception is found, the laws will be modified to include the 
exception. The laws have some background assumptions and require careful defini- 
tions. Chemical thermodynamics is dominated by the concept of equilibrium. Equi- 
librium is a state or condition that remains unchanged as long a.s energy and matter 
are neither gained nor lost from the system. An equilibrium system will return to 
equilibrium after a slight perturbation, such as a change in temperature or pressure. 

Equilibrium principles require that a system be completely described by easily 
measured variables, such as temperature, pressure, and composition. These proper- 
ties of the system define its slate. The state is relatively easy to define for simple sys- 
tems in the chemical laboratory, but is much more difficult to define in the complex 
systems of nature and soils. For example, the complete mineralogical composition 
of soils is difficult, if not impossible, to determine. The amounts and types of min- 
erals in the clay fraction are known at best only semiquantitative^. In addition, the 
composition of each mineral can vary over a wide range and the weathered surfaces 
are different than the unweathered portion. Soil behavior also depends on the size, 
matrix, and interactions of soil particles. These are at present inadequately measured 
or defined. Wetting and drying can irreversibly change the arrangement of soil parti- 
cles. Heating and drying can destroy organic compounds, change soil minerals, and 
markedly modify the composition of the soil solution. 

Soils are nonequilibrium systems, but sometimes can be considered close enough 
to equilibrium to let equilibrium principles apply. The requirements are that (1) the 
rates of soil change are negligible in the time scale under consideration, (2) the per- 
turbation does not change the composition of the solid phase, (3) the soil is homo- 
geneous, and (4) the soil is a closed system — it does not gain or lose matter — during 
the experiment. None of these assumptions can be completely true, but the errors can 
be small if the experiments are done carefully. 

The "laws" of thermodynamics describe physical and chemical behavior to which 
we have not yet found exception. Claimed exceptions to the laws of thermodynam- 
ics have thus far proved to be unrecognized, and often very subtle, violations of 
the conventions of thermodynamics. The first law — matter and energy are conserved 
(neither created nor destroyed) during a process — is an idea that already appealed to 
medieval philosophers. During the late 1700s and early 1 800s, careful measurements 
of water pumps and steam engines showed, however, thai although mass was con- 
served, some energy was always lost. Furthermore, the faster the pumps and engines 
ran, the greater the loss of this energy; that is, the process was less efficient at greater 
speeds. 

Rather than discard the first law despite this apparent flaw, people changed the 
definitions, and introduced the second Jaw, so the first law would still be true. The 
first law now deals with total energy, which is conserved. 

If a reaction proceeded infinitely slowly, the first law would still hold true. Real- 
world reactions, however, proceed at a finite pace. The second law takes this into 
account by stating that some energy, called entropy S, is irretrievably lost during any 
process. This brings thermodynamics much closer to the real world. The available 



THERMODYNAMICS 89 

energy (Gibbsfree energy G) obtained from a process is always less than the energy 
input H (enthalpy) or first law energy 

AG = AH - T AS (3.32) 

where T is the absolute temperature and the A refers to a difference between two 
values. We do not know the absolute energies; we can only measure the difference in 
energy between two states. The terms in Eq. 3.32 are given positive or negative signs 
according to the change within the system. The system might be a volume of soil or 
a flask containing a soil suspension, considered as separate from its surroundings. If 
the system loses energy such as heat or work to the surroundings, the sign is negative 
because the system has less energy than before. If the system gains energy from the 
surroundings, the sign of the energy terms is positive. 

Entropy is defined so that it is positive and increasing during spontaneous reac- 
tions, reactions that proceed without energy input. Entropy is the energy lost during 
the reaction. This energy ultimately is radiated into space and is made up for by solar 
energy plus some energy from the earth's interior. Without solar and earth energy 
to compensate for the continual loss of entropy, the earth would eventually reach 
equilibrium and life would end. Some people consider entropy to be the unavailable 
energy or a bookkeeping entry that corrects the first law. Others prefer to describe 
entropy in more physical terms, as the friction in all processes, or as the energy lost 
in rearranging ions and molecules during chemical reactions. Increasing entropy im- 
plies increasing randomness or disorder of matter; decreasing entropy implies struc- 
tural ordering. 

Soon a loophole was discovered in the second law. When the temperature is ab- 
solute zero, the second law reduces to the first law and becomes unnecessary. So the 
third law — that we cannot reach absolute zero — was invoked to keep the second law 
universally valid. Corollaries of the laws are that entropy strives toward a maximum, 
free energy strives toward a minimum, and when the free energy is at a minimum, 
the system is at equilibrium. 

Systems far from equilibrium, such as living systems that are negative entropy be- 
cause they are so ordered, may be better described by irreversible thermodynamics. 
Living systems are greatly ordered and generally have active energy flows; examples 
are ecological systems having organisms with elaborate organic compounds closely 
interacting with other compounds, active population growth, and photosynthesis. Ir- 
reversible thermodynamics attempts to describe this energy flow. Irreversible ther- 
modynamics has found little application in soil chemistry. In many cases, soil min- 
erals and solutions are close enough to equilibrium that equilibrium and reversible 
thermodynamics suffice. Often a soil process in the laboratory or field is viewed as 
reacting much more rapidly than the overall soil. Then the process can be treated 
essentially as if it were an equilibrium process. 

The three laws of thermodynamics and Newton's laws of physics govern the be- 
havior of matter and energy in the systems that we normally deal with. The 20th- 
century developments of quantum and statistical mechanics deal with atomic and 
subatomic behavior. One test of their validity is that they yield the laws of thermo- 



90 WATER AND SOLUTIONS 

dynamics when expanded to larger systems. So far, soil behavior has not required 
quantum or statistical mechanics for explanation. 

Soils are sensitive to changes brought about by drying, wetting, leaching with 
monovalent salt solutions, acidification and basification, changing ionic strength, 
changing oxidizing-reducing conditions, and changing the soil-solution ratio. The 
bulk of the soil is quite resistant to change, but soil surfaces are vulnerable. Some of 
these reactions revert back slowly if at all to the previous state. This irreversibility 
weakens assumptions of equilibrium. Experiments that hope to utilize equilibrium 
principles should not inadvertently change the chemistry of the soil's surface. 

A3.1.1 Gibbs Free Energy 

The thermodynamic term of widest use in soil chemistry is the free energy, or more 
explicitly, the Gibbs free energy. This is the energy of a substance or a reaction that, 
at constant temperature and pressure, is available for subsequent use. Energy drives 
chemical reactions and AG is the most widely useful. It is directly related to (1) the 
activity or chemical potential, (2) the energy of formation of compounds, (3) the 
equilibrium constant of a reaction, and (4) the electrode potential. The first three are 
discussed here; the electrode potential is discussed in Chapter 4. 
The AG can be defined as 

AG = V AP -t- S AT + /(compositional, electrical, gravitational potentials) 

(3.33) 

where V is the volume, A/ 3 the pressure change, and AT the temperature change. 
At constant temperature and pressure 

AGt.p = /(compositional, electrical, gravitational potentials) (3.34) 

The Gibbs free energy is determined by changes in composition, gravity, and elec- 
trical potentials. Chemistry is usually concerned only with variations in composition 
potentials. The gravitational potential arises from differences in elevation, such as 
the "water head" or hydraulic potential of soil physics, but can usually be ignored 
by soil chemists. Electrical potential is an important consideration near charged sur- 
faces such as soil particles. When dealing with an aqueous solution, however, the 
Gibbs free energy at constant temperature and pressure is determined solely by the 
composition and concentration of the solution. 

The most useful concentration unit for solutions is the chemical activity a (Sec- 
tion 3.2). The change in tree energy with the amount of solute is 

AG = RT \na = 5.71 loga (3.35) 

where the units of AG are kJ mol - ' . The free energy decreases as the solute becomes 
more dilute. The change is calculated from an arbitrary standard state of solute, usu- 
ally defined as an ideal 1 M solution, where a = 1 . At solute activities up to unity, the 
activity of the solvent (water) is usually assumed to be unity, that is, to be unaffected 
by solute concentration. 



THERMODYNAMICS 91 

The AG value of a certain state is the difference in free energy between that state 
and a standard state. For the free energy possessed by a chemical compound, the 
usual standard state is its free energy at 25° C (298. 1 5 K) and 1 5 Pa ( 1 atm) pressure. 
The elements are assigned free energies of zero. The energy of formation is absorbed 
or released by compounds when they form from their elements. For example, careful 
measurement of the reaction 

H 2 + \0 2 = H 2 (3.36) 

at standard conditions (298.15 K and 10 5 Pa) yields 237 kJ of available energy per 
mole of water formed. Thus, AG° — —237 kJ mop' with the superscript denoting 
standard conditions. This reaction releases energy to its surroundings, so the sign of 
AG° is negative, indicating that the system has less energy than before. 

Reactions that release energy and leave the system in a lower energy state, a more 
stable state, than before are spontaneous. The second law states that systems will 
strive to reach the lowest energy level. Thermodynamics says only that a reaction 
will proceed and not what the reaction rate will be. The hydrogen and oxygen in 
reaction 3.36 can coexist for centuries without reacting until a catalyst or spark is 
introduced. 

The free energies of formation of many compounds and ions have been measured 
and compiled. Table 3.7 contains values for the AC of formation of some com- 
pounds relevant to soil chemistry. 

The change of free energy during a reaction is the difference between the free 
energies of the products and those of the reactants. An important energy reaction in 
nature is photosynthesis, the formation of glucose: 

6C0 2 + 6H2O = C 6 H| 2 6 (glucose) + 6O2 AG° = 2879 kJ moP 1 (3.37) 

The sign of the free energy change of this reaction is positive, because solar energy 
has come from the surroundings and been trapped in the glucose molecule. The free 
energy of the reaction is the difference between the free energies of formation of the 
products (glucose and oxygen) and of the reactants (carbon dioxide and water): 

AG3.37 = AG g | ucose 4- 6(AG 02 ) - 6(AG c o 2 + AGh 2 o) (3-38) 

Each term is the value per mole so it is multiplied by the appropriate coefficient from 
Eq. 3.37. Rearrangement yields 

AG„ lucosc = AGa.yj - 6(AGo 2 ) + 6(AG c o 2 + AG H2 o) (3.39) 

The AG of formation of oxygen is assigned a value of zero, because it is a pure 
element. The AG of formation of carbon dioxide (-394.3 kJ moP ' ) and other com- 
pounds is available from many handbooks. The AG of formation of glucose is thus 

AG n i UCOS e = +2879 - + 6(-394.3) + 6(-237.2) = -910.4 kJ moP 1 (3.40) 



92 



WATER AND SOLUTIONS 



Table 3.7. Standard Gibbs free energies of selected compounds at 25° C (a is activity 
and P is partial pressure or mole fraction of gas) 



Formula 


Name or State 


AG°(kJinor') 


Source 


Al(OH) 3 


Gibbsite 


-1 151 


3 


Al 2 Si 2 5 (OH)4 


Kaolinite 


-3 783 


3 


M 0.56 ( A1 3.03 M g0.58 Fe 0.45 


Montmorillonile 


-10330 


4 


(Si 7 .87Al0.13)O20(OH) 4 








CaC0 3 


Calcite 


-1 129 


I 


CaS0 4 • 2H 2 


Gypsum 


-1797 


1 


Fe 2 3 


Hematite 


-741 


2 


Fe(OH) 3 


(Crystalline?) 


-694.5 


2 


FeC0 3 


Siderite 


-674.0 


2 


FeS 2 


Pyrite 


-150.6 


2 


MgC0 3 


Magnesite 


-1029 


2 


Mn0 2 


Pyrolusite 


-464.8 


2 


MnC0 3 


Rhodochrosite 


-817.6 


2 


MnS 


Alabandite 


-233 


2 


HN0 3 and NO~ 


a= 1 


-110.6 




NH 3 


P= 1 


-16.64 




NH 4 OH 


a = 1 


-263.8 




11+ 


a= 1 


0.0 




OH- 


a= 1 


-157.1 


2 


H 2 


a= 1 


-237.2 




H 3 P0 4 


a = 1 


-I 143 




po^- 


a = 1 


-1019 




K+ 


u= 1 


-282.0 


2 


KAlSi 3 Og 


Feldspar 


-3 581 


2 


KAl 3 Si 3 O )0 (OH) 2 


Muscovite 


-5 558 


3 


Si0 2 


Quartz 


-8567 




Si(OH) 4 


Soluble silica, a = 1 


-1317 




S0 2 


P = 1 


-300.2 




H 2 S 


P = 1 


-33.6 




H 2 S0 4 and SO^" 


a = 1 


-744.6 




Ti0 2 


Rutile 


-888.2 


2 


C0 2 


P = 1 


-394.3 




CH 4 


P = 1 


-50.75 




H 2 C0 3 


a = 1 


-623.2 





1. From D. D. Wagman ct al. 1968. Selected values of chemical thermodynamic properties. U.S. Bur. 
Standards Tech. Notes 270-3 and -6, Washington. DC. 

2. From R. M. Garrels. I960. Mineral Equilibria. Harper, New York. 

3. From S. V. Mattigod. 1976. Ph.D. Dissertation. Washington State University, Pullman. 
4..1. A. Kiltrick. 1 97 1 . Soil Sci. Sot:. Am. Proc. 35:140. 



SOLI D SOLUTIONS AND OPEN SYSTEMS 93 

When reaction 3.37 reverses during respiration, 1 mol of glucose liberates —2879 kJ 
of available energy to fuel the life processes of living organisms. 

A third important facet of the Gibbs free energy is its relation to the equilibrium 
constant of a reaction. A reaction proceeds until the components are at their lowest 
energy level, the most stable state. This state is defined by the equilibrium constant 
K: 

AC = RT la K= 5.71 log K (3.41) 

Equation 3.41 requires that the standard states of the products and reactants be 
known, that the components can be defined quantitively and in a thermodynamic 
sense. In soils and much of nature these definitions are rarely possible. The states 
of ions or molecules in soil systems, and in probably all colloidal systems, are ill- 
defined thermodynamically. In rigorous thermodynamic terms even ions are unde- 
fined. Soil reactions, because of the nonequilibrium in soils and the lack of defined 
standard states, yield reaction coefficients, rather than reaction constants, and their 
values vary with soil conditions. 



APPENDIX 3.2 SOLID SOLUTIONS AND OPEN SYSTEMS 

The thermodynamics of solid mixtures and solid solutions would seem to hold great 
promise for soil chemistry, as much as the thermodynamics of aqueous solutions 
has proved useful, but it has been largely neglected. The neglect of solid solutions 
is partly due to an incompatibility between classical thermodynamics and nature. 
Nature and soils are more complex and they are open systems — energy and matter 
flow in and out of the system being studied. Nonetheless, some of the principles of 
solid mixtures are tacitly applied to ion exchange equations and adsoiption studies. 
Direct applications of the thermodynamics of mixtures to soils may have consider- 
able merit. The necessary assumptions are similar to those that have already been 
accepted in adsorption and cation exchange studies. 

Strictly speaking, thermodynamics applies only to total equilibrium and to closed 
systems. A closed system gains and loses no matter during the reaction. Soils steadily 
lose matter during weathering and gain matter by aerial deposition, rain, and tectonic 
movements. Soils also are not at total equilibrium; the world would be sterile if nature 
were at total equilibrium. 

Soil chemistry cannot afford the luxury of rigorous thermodynamics and instead 
has to stretch and bend the rules into what is called exlrathermodynamics, but the 
bending and stretching must be done wisely. Cation exchange, solubility, Donnan, 
and adsorption studies find use in soils and assume equilibrium and a closed system 
during the experiment. Applying the thermodynamics of solid solutions to the reac- 
tion between the soil solution and soil particles requires the same rule bending as the 
other studies. 

When substances in liquids or solids mix completely on an atomic scale, that is, 
when they mix homogeneously and randomly, the mixing usually decreases the es- 



94 WATER AND SOLUTIONS 

caping tendency of the components. The escape can be in the form of molecules 
evaporating from a liquid mixture, or of Ca and Mg ions dissolving into water from 
dolomite CaMg(CX>?)2- Other examples of escaping tendency are cations exchang- 
ing from the mixture of cations on soil surfaces, phosphate desorbing from soil 
solids, and trace ions dissolving from silicate minerals. 

Ions in solids cannot mix as randomly as can the components of gases and liquids. 
Mixing in glasses is close to random; ions in crystals are mixed but confined to fixed 
positions. This degree of mixing in crystals is sufficient to allow the thermodynamics 
of mixing to apply, and the mixing greatly affects the aqueous solubility of the ions in 
these solids. The general name of these solid mixtures is solid solutions; Hildebrand 
named crystalline solid solutions with their more limited mixing as regular solutions; 
the mixing of ions in solids is regular and repeated. 

One example of the effect of mixing on the free energy or escaping tendency is 
the evaporation of a mixture of two organic liquids. When hexane is mixed with 
heptane, hexane's escaping tendency decreases. The escaping tendency is measured 
by its partial pressure P, which depends on hexane's volatility in the pure state and 
its concentration in the mixture, its degree of mixing: 

^hexane = gAjChexane (3.42) 

where Po is the vapor pressure of pure hexane, C is its mole fraction concentration 
in the mixture, and g is the activity coefficient that accounts for deviation from ideal 
mixing. Mixing increases the randomness of the hexane molecules, increases hex- 
ane's entropy, and therefore decreases its free energy, escaping tendency, and vapor 
pressure. Alternatively, the probability of a hexane molecule evaporating from a mix- 
ture is less than if evaporating from pure hexane, because fewer hexane molecules 
are at the surface. 

Similarly, the escaping tendency, or aqueous solubility, of an ion depends on its 
bonding strength to the other ions of the solid. For ions on the surface, the escaping 
tendency depends on the extent of its mixing on the surface. As an example of the 
effect of solid mixing, the aqueous solubility of AIPO4 is expressed by AlP04's 
solubility product when the mineral is pure: 

(awpoo (343) 

«AIP0 4 

For pure AIPO4, the activity a — I. When the phosphate ions are instead adsorbed on 
the surface of Al(OH)3, the phosphate ions behave as if they are AIPO4 mixed with 
Al(OH)3. The solubility of AIPO4 expressed by its ion activity product (AIXPO4) in 
the aqueous phase will be less than its solubility product because mixing increases 
the entropy of AIPO4 ' n the solid phase and reduces its aqueous solubility. Its solu- 
bility in the mixture on the surface is 

IAP A ,ro 4 = #tfs P C A iP0 4 OM) 



SOLID SOLUTIONS AND OPEN SYSTEMS 95 



Q. 

3 




0.4 0.6 0.8 
Mole fraction, X 

FIGURE 3.6. Change of relative equilibrium aqueous solubility, \AP/K sp , of a component in a 
solid solution as a function of its composition in the solid. The solid solution is assumed to be 
ideal. (From H. L. Bohn. 1992. Soil Sci. 164:357.) 



where g is the activity coefficient, A" sp is the solubility product of pure AIPO4, and 
C is the mole fraction of phosphate on the Al(OH>3 surface. 

Figure 3.6 shows the effect of this solid solution mixing on the aqueous solubility 
of a substance assuming ideal mixing, g = 1 . Mixing has little effect on the 1AP/K sp 
ratio of a substance until its mole fraction is <0.5. The reduced aqueous solubility 
due to this mixing should be pronounced for trace ions in the soil solution such as 
phosphate and the trace melals, but insignificant for Si, Al, and Fe. 

The activity coefficient g is an index of the deviation from ideal mixing. In ideal 
mixing, only the mixing entropy affects the escaping tendency. Mixing in nature 
is nonideal; molecular and ionic interactions change the amount of interaction and 
therefore affect the activity coefficient. The activity coefficients of solutes in water 
are generally < 1 because ions interact with each other so their effective concentration 
is less than their actual concentration. Activity coefficients of solid components are 
usually > 1 because substituted ions usually do not fit easily into the solid's structure. 
The structural stresses imposed by substitution tend to expel the ions into the aqueous 
solution. This somewhat counteracts the decreased aqueous solubility caused by the 
entropy of mixing and yields activity coefficients > 1 . Each ion is different and fits 
differently into soil mineral structures. The usual result is that the ion's solubility 
is greater than if it mixed easily into the structure, hence g > 1, but not enough to 
overcome the effects of the mixing entropy. Early measurements indicate that the g 
values for ions like Al that ate common in soil minerals, and presumably substitute 
easily, are about g = 3 to 5. For ions like phosphate and Ca that do not substitute 
easily into aluminosilicales, g > 20. 

Like activity coefficients in aqueous solutions, solid activity coefficients are con- 
centration dependent. As more phosphate ions mix on the surface of Al(OH)3, for 
example, the hydroxide structure becomes more and more strained. The strain is re- 



96 WATER AND SOLUTIONS 

fleeted in a positive deviation from ideal mixing. PQ4 ions mix Jess easily because 
other PO4 ions are already on the surface, and the activity coefficient increases in 
value. The Al(OH) 3 tries to release the strain by trying to expel phosphate from the 
surface. As more phosphate is added, AIPO4 eventually becomes more stable than 
the mixture and AIPO4 precipitates as a separate phase. 

Solid activities have been ignored largely because of a misunderstanding in our 
early chemistry training. The solubility product of Al hydroxide in chemistry texts, 
for example, is written as 

K, p = (AI)(OH) 3 (3.45) 

Equation 3.45 is accurate only in a pure system that contains only Al and OH ions 
and water. In that system the activity of Al hydroxide is one. For systems containing 
other components, such as aluminosilicales and Al substituted into FeOOH, a more 
complete expression of Al solubility is 

(Al)(OH)3 
Al(OH) 3 solid 

and 

IAPakoh);, = £^spCakoh)j (3.47) 

The aqueous solubility of Al can be related to the solubility of Al(OH) 3 even though 
that mineral may not be present. The solid activity of Al(OH) 3 is gC. The aqueous 
solubility of Al depends on the equivalent concentration of Al(OH) 3 in the solid 
phase. The activity of the solid is defined by the activity of the ions in the aqueous 
solution. The TAP is a measurable property of the system, so the solid activity is a 
thermodynamic property. 

The concentration of Al hydroxide in soils is usually high enough that Eq. 3.45 
is adequate and Eq. 3.46 is unnecessary. That is probably not true for ions in trace 
concentrations in soils. The mole fraction concentrations of phosphate and transition 
metal ions in soils are <§;0.01. So soil phosphate solubility at "equilibrium" can be 
orders of magnitude less than the solubility of pure Al phosphate, as many soil mea- 
surements show. We can call it equilibrium because the solubility changes slowly to 
imperceptibly with time. If anything, the aqueous phosphate concentration decreases 
with time as the surface Al ions slowly diffuse into the weathered surface, mix further 
with other ions, and further increase (heir entropy. Simple AIPO4 solubility product 
equivalent to Eq. 3.45, on the other hand, predicts that the phosphate concentration 
will increase with time. 

Mixing on soil surfaces explains some soil phenomena very well. Mixing explains 
why soils can adsorb virtually every ion from soil solutions and can retain those ions 
much more tightly than can the ion's own hydroxyoxides, or retain phosphate more 
tightly than Al phosphate. This happens even though soil minerals are continuously 
losing their own components by weathering. Soil "adsorption sites" are areas where 
ions from the soil solution can mix with ions on soil surfaces. The mixing ability 



KINETICS 97 

varies with the kind of ions on the soil's surfaces and with the ability of the aqueous 
ion to bond strongly with those surface ions. 

Soil is an open system. Weathering carries away substances from the soil; wind 
and rain add others. The surfaces of soil particles reflect these processes as well as the 
composition of the internal part of the particles. Because the particle and its surface 
do not have the same composition, the particle is not at equilibrium with itself. Ion 
diffusion within crystals is slow enough that this disequilibrium can be ignored. 

In the weathered surfaces of soil particles where some semiliquid water may be 
present, ion diffusion may be fast enough to affect laboratory and field experiments. 
The slow removal of phosphate ions by soils from the aqueous phase may be due 
to the slow diffusion of PO4 to Al and Fe ions within the weathered surface. The 
increasing strength of trace metal retention by soils with time may similarly be due 
to such diffusion into the semisolid, weathered surfaces of soil particles. 

Ion activities in the soil solution can also be treated as being governed by the 
saturation index of a mineral: 

IAP 
Saturation Index = — — (3.48) 

From Eqs. 3.47 and 3.48, the saturation index equals the mineral's solid activity 
coefficient times its mole fraction in the solid phase: 

Saturation Indexj = g\C\ (3.49) 



APPENDIX 3.3 KINETICS 

Thermodynamics predicts that substances will react until they reach their most stable 
states, but does not say how the most stable state will be achieved or how long it will 
take. Not all reactions lead immediately to the most stable states, and some reactions 
are exceedingly slow. Kinetics is the study of these reaction mechanisms: the rates, 
paths, and intermediate products of chemical reactions. 

A substance put in conditions in which it is unstable will sometimes not react at 
all. A mechanical example of such "metastable equilibrium" is a rectangular block 
standing on end. It will not reach the more stable state of lying on its side until it is 
pushed and lifted so that its center of gravity is beyond its edge. 

In chemical reactions, pushing and lifting the block correspond to activation en- 
ergy. A mixture of H2 and O2 gases will not react until a spark or high temperature 
provides sufficient activation energy to greatly perturb the metastable equilibrium 
and allow the gases to react. Figure 3.7 shows the change in energy of a mixture 
of substances A and B that requires some activation energy to create the activated 
state AB* before it degrades to C and D. Photosynthesis is an example of this pro- 
cess. Sunlight provides the activation energy that creates an activated state, glucose, 
which is metastable. The activated state returns to the stable initial states, CO2 and 
H2O, through a path of metabolism and decay that is as intricate as photosynthesis. 



98 WATER AND SOLUTIONS 



>- 
UJ 

z 



A* B / 


AB* 


Tactivation 

l 1 ENERGY 




1 OD 





REACTION DIRECTION ► 

FIGURE 3.7. A representation of the extra energy (activation energy) needed to carry out the 
reaction A + B -► C + D. AB* is the intermediate, activated complex. 



Nitrogen fixation, denitrification, soil weathering, phosphate fixation, clay min- 
eral degradation, and potassium and transition metal fixation are problems for which 
the reaction rates are usually as, or more, important than equilibrium. Most soil 
chemical applications of kinetics have been in soil microbiology and soil biochem- 
istry, where the lack of equilibrium is more obvious. The use of kinetics in inorganic 
soil chemistry will undoubtedly broaden in the future. It can even be argued that 
kinetics is basic to thermodynamics, because equilibrium is the condition where op- 
posing reaction rates are equal. 

Small amounts of some substances can increase reaction rates enormously. These 
substances, when left unchanged by the reaction, are called catalysts. Perhaps the 
simplest type of catalytic action occurs when a surface adsorbs the reactants so that 
they remain in close proximity for relatively long periods of time. The probability of 
forming a new compound from the reactants then is much greater than if they merely 
collide and rebound in a gaseous or liquid phase. Soil surfaces may act as catalysts 
in this way. 

Catalysts lower the activation energy barriers that hinder reactions. The activa- 
tion energy requirement can arise from many chemical and physical factors, and the 
mechanisms by which catalysts lower the activation energy are probably just as nu- 
merous. Iron, manganese, and other transition-metal ions catalyze electron transfers 
during oxidation-reduction reactions. Enzymes are the organic catalysts indispens- 
able for most reactions of living organisms. 

Reaction inhibitors slow reaction rates. Nitrogen mineralization and nitrification 
(conversion of organic nitrogen and ammonium to nitrate) rates in soils, for example, 
can be slowed temporarily by chemicals that specifically slow or stop the microor- 
ganisms involved. Toxic metals can also operate as enzyme inhibitors, by replacing 
the metal coenzyme portion of an enzyme and thereby inactivating it. 

Kinetics is being employed increasingly to study the soil chemistry of carbon, 
nitrogen, potassium, phosphate, and trace metals. The soil reactions of these elements 
are often slow enough to be experimentally measurable. Because carbon and nitrogen 
cycle back and forth between soil, water, plants, animals, and atmosphere faster than 



KINETICS 



99 



the rates at which they reach their most stable thermodynamic states, kinetics is an 
appropriate tool with which to investigate their behavior. 

For inorganic ions, the reactions themselves can be very fast, but the ions may 
have to diffuse through soil pores before they reach a reaction site. The ions may 
also have to diffuse through the weathered surface. Diffusion processes lend them- 
selves to kinetic treatment. With multiple diffusion and reaction processes going on 
simultaneously, the kinetic treatment can become very complex. 

A3.3.1 Reaction Order and Rate Constants 

One approach of kinetics is to describe the dependence of reaction rates on reactant 
concentrations. For instance, the rate of phosphate fixation depends at least partly 
on the amount of fertilizer added, and the rate of denitrification (the conversion of 
soil nitrogen, usually nitrate to N-2 and N2O) depends on soil solution nitrate concen- 
trations. Kinetics relates reaction rates and reactant concentrations by means of the 
reaction order and the reaction rate constant. The denitrification rate (- ANO3 /Af) 
is presumably related to soil nitrate concentration by 



ANOJ 

a7~ 



= ^A{N 2 0}"=fe'{soilNO^}" 



(3.50) 



where k and k' are the reaction rate constants or coefficients, n is the reaction order, 
and braces denote concentrations. The negative sign indicates that nitrate is disap- 
pearing, and the positive sign indicates the production of Nt and N2O gases. 

In pure systems under carefully controlled conditions, the reaction order is often 
0, 1 , or 2. These reaction rates can be plotted linearly with respect to time by choosing 
the appropriate concentration axis (Fig. 3.8). The reaction order can be obtained by 
fitting data to such plots. Zero-order reaction rates are independent of the amount or 
concentration of the reactant studied: 



AC 
~Kt~ 



= Jfc 



(3.51) 



where C is the concentration of some substance that is disappearing at a rate that is 
constant with time. 




logC 




l/C 



Time 



Time 



Time 



FIGURE 3.8. Zero-, first-, and second-order reactions plotted linearly with time. Note the varying 
units of the vertical axis. 



100 



WATER AND SOLUTIONS 



The most common order found is first order, in which the reaction rate depends 
on the concentration of one reactant A: 



A[A] 
A/ 



= *[A][B] 



(3.52) 



This order can be obtained by "swamping" the system with the other components so 
that reactant A is rate limiting. 

Second-order reactions depend on the concentrations of two reactants, or on the 
concentration of one reactant squared. The reaction of A and B to form D, for exam- 
ple, might follow the equation 



AD 

~a7 



= *[A][B] 



(3.53) 



The sign of the left side is positive because D is increasing. If the concentration of 
B were much higher than that of A, the reaction rate would appear to depend solely 
on A. The reaction rate would be pseudo-first-order with respect to A and almost 
pseudo-zero-order with respect to B. The order of a reaction therefore depends on 
the conditions of the experiment. 

The rates of soil adsorption reactions may also depend on the exponential of the 
amount already adsorbed. Phosphate adsorption by soils, for example, sometimes 
follows the Elovich equation: 



AA 



ads 



Ar 



<*e ''Aads 



(3.54) 



where A a ds is the amount already adsorbed and a and ft are empirical constants. Be- 
cause of its two constants, this equation is more easily fitted to experimental data. 
The equation is plotted in Fig. 3.9, where a is the slope of the line. Reaction mecha- 



z 
O 

a: 

t- 
z 
w 
o 
z 
O 
o 

z 
O 



o 



In t 



FIGURE 3.9. Plot of the Elovich adsorption equation 3.54. 



KINETICS 101 

nisms sometimes can be inferred from measured reaction orders. After ensuring that 
no other reactions are rate limiting, the reaction mechanisms are inferred by split- 
ting the reaction into hypothetical step reactions until the sum of the orders equals 
the measured order. Insuring that the reaction in question limits the reaction rate is 
difficult for soils. The number of steps between reactant and product can be very 
large. 

Reaction rates in soils are more complex than those in pure systems. Secondary 
and side reactions are difficult, if not impossible, to control. The measured reaction 
order is therefore usually fractional rather than a whole number because other re- 
actions are usually going on at rates different from the one in question. If only the 
total change of a component is measurable, the overall reaction rate and order are 
weighted according to the relative contribution of each reaction. 

In the denitrification example (Eq. 3.50), the rate of nitrate loss depends also on 
microbial activity and therefore on the presence of an energy source: 

^ = ft [soil NO"]" [available CJ'" (3.55) 

The reaction order is the sum of n plus m. Competing reactions of different rates, and 
reverse reactions, are the rule in nature rather than the exception. Reaction orders of 
such complex systems are usually nonintegral. Laboratory or field measurements 
are often possible only if all variables except the one of interest are held constant. 
Because the effects of only one or a few variables are measured, the reaction rate is 
incompletely described and the order is actually a pseudo-order. The term "pseudo" 
unfortunately has a disparaging connotation. Here it implies only that the system is 
too complicated to measure completely. 

Determining the rate constant and order of a reaction is tedious and time- 
consuming. For many studies, this detail is unwarranted and the half-life is measured 
instead. The half-life is the time required for half of the original concentration of 
reactant to disappear. For the particular case of a first-order reaction, the half-life 
(?i/2) is directly related to the reaction rate constant ft by 

In Cq/2 0.693 
«i/2 = - ir - = — (3-56) 

where Co is the original reactant concentration. Because many soil studies are carried 
out under conditions involving only one experimental variable, the pseudo-orders of 
the reactions may be close to unity. In these cases the half-life is a useful semiquan- 
titative indicator of reaction rate. 



A3.3.2 Temperature Effects 

Higher temperatures increase the energy and probability of particle collisions. These, 
in turn, generally increase reaction rates. Measuring reaction rates at various temper- 
atures can provide useful clues about reaction mechanisms. Figure 3.10 shows the 
effect of temperature on the rates of three types of reactions. The exponential rise 



102 



WATER AND SOLUTIONS 




Temperature 

FIGURE 3.1 0. Typical effects of temperature on reaction rates of (A) inorganic reactions, (B) ex- 
plosive reactions, and (C) biochemical reactions. 



of curve A is characteristic of inorganic reactions, the rate increases two- or three- 
fold with each 10° C value, or the increase in reaction rate with a 10° C increase 
in temperature, gio — 2 to 3. The plot of log* versus ]/T is often linear for such 
reactions. Such data can yield the activation energy E d from the Arrhenius equation: 



A log* _ £ a 
A7' ~ 2.303RT 2 



(3.57) 



Another response of rate to temperature is an explosion, curve B of Fig. 3.10. The 
reaction rate increases only slowly until a critical temperature is reached, whereupon 
the rate approaches infinity. 



A3.3.3 Microbially Catalyzed Reactions 

Many chemical reactions in soils occur at measurable rates only because of enzy- 
matic or microbial catalysis . Curve C of Fig. 3 . 1 is characteristic of biological reac- 
tions. Biological reactions generally increase about threefold per 10° C rise in tem- 
perature ( Q io = 3) up to an optimum temperature and then decrease rapidly at higher 
temperatures. Most soil organisms are mesophiles, whose optimal temperatures are 
30 to 37° C; 37° C is particularly common. Soils also contain thermophites, whose 
optimum temperatures are 55 to 60° C, and psychrophiles, whose optimal range is 5 
to 15° C. Psychrophiles were once thought to be rare in soils, but recent work sug- 
gests that their incidence and importance has been underestimated. The mean annual 
temperature of most soils is in the 5 to 15° C range. 

When biochemical reactions are studied with time, two situations commonly oc- 
cur. One is when enzyme concentrations remain constant (the biological population 



KINETICS 



103 



in effect remains constant), and Michaelis-Menten kinetics often apply. This treat- 
ment assumes that the enzyme (E) and the reactant or substrate (S) form a complex 
(ES) that dissociates either to the original substrate and the enzyme, or to a new 
product (P) and the enzyme. Thus, 



S + E?=±ES 4- E + P 

*2 



(3.58) 



Michaelis and Menten found that the rate of substrate disappearance with time could 
be expressed as 



A[S] fc 3 [E][S] 



At K m + [S] 



(3.59) 



where the brackets indicate concentrations and where K m is the Michaelis constant. 
Equation 3.59 describes the curve of Fig. 3.1 la. Equation 3.59 assumes that all non- 
specified rate factors are at steady state during the period of measurement. 

When the substrate concentration is much smaller than K m , the denominator of 
Eq. 3.59 reduces to K m and the reaction rate, at constant enzyme concentration, is 
first order with respect to the substrate concentration \S]: 



*3 



A[S] = 

At K m [E)[S] 



= k'[S] 



(3.60) 



This equation describes the steep portion of the curve near the origin in Fig. 3.1 la. 
When [S] is much larger than K m , Eq. 3.60 reduces to 



A[S] 
At 



= *3[E] 



(3.61) 




Rate Factor (light, nutrient, etc.) 



Time 



FIGURE 3.11. Change of reaction rate (A) with a rate factor according to the Michaelis-Menten 
kinetics and (B) with time and population response. 



1 04 WATER AND SOLUTIONS 

which means that the rate is independent of, or zero order with respect to, S. The rate 
then depends only on the steady-state concentration of the enzyme, that is, on the 
plant or microbial population. 

The rate of an enzymatic reaction depends on many factors. Increasing one fac- 
tor generally increases the rate less than the increased amount of the factor, because 
the rate is now hindered by other growth factors. For instance, nitrogen fertiliza- 
tion may increase crop growth rates, but plants must still contend with limited water 
and nutrient supplies, disease, and climate. Each successive increment of added ni- 
trogen will have less effect on growth, because other factors are now rate limiting. 
Only by changing other rate-limiting conditions can an organism take better advan- 
tage of an added growth factor. Michaelis-Menten kinetics apply both to short-term, 
single-plant experiments and to the worldwide plant response to changing PrjOi m 
the atmosphere. In both cases the population and enzyme concentrations remain con- 
stant. 

The second situation is more common in experimental work; the biological pop- 
ulation increases during the study period in response to some growth factor. This re- 
quires time and causes the characteristic lag time before biological reactions begin. 
This lag distinguishes reactions that are primarily biochemical. Inorganic reactions, 
in contrast, are usually most rapid at the beginning of the experiment. 

During the lag period, the growth of the organism and the corresponding reaction 
rate are slow but increase exponentially with time. Later the rate slows to zero when 
a new set of limitations is reached. This results in the familiar S-shaped or sigmoid 
curve (Fig. 3.1 lb). The curve is described by equations such as 

AN A[S] rN(N iim -N) 

— = (3.62) 

Af At N lim 

where N is the number of organisms, N\\ m is the maximum population of organisms, 
and r is the difference between the organism's coefficients of birth and death rates (or 
forward and backward reaction rates). Equation 3.62 applies to such situations as the 
soil's evolution of CO? from freshly introduced organic matter, and the appearance 
of nitrate in soils recently fertilized with ammonium salts. 

The application of kinetics and thermodynamics requires a deeper understand- 
ing than the brief introduction given here. Although best suited to simple systems, 
thermodynamics and kinetics are also unexcelled as tools for the understanding of 
chemical phenomena in nature. 

BIBLIOGRAPHY 

Bockris, J. O., and A. K. N. Reddy. 1970. Modern Electrochemistry, Vol. I. Plenum/Rosetta, 

New York. 
Bohn, H. L. 1992. Chemical activity and aqueous solubility of soil and solid solutions. Soil 

Sci. 154:357-365. 
Lindsay, W. L. 1979. Chemical Equilibria in Soils. Wiley-Interscience, New York. 
Sposito, G. 1981. The Thermodynamics of Soil Solutions. Clarendon Press, Oxford, UK. 
Slumm, W., and J. J. Morgan. 1995. Aquatic Chemistry, 3d ed. Wiley, New York. 



QUESTIONS AND PROBLEMS 1 05 

QUESTIONS AND PROBLEMS 

1. What is the pH of 0.1 M HC1, 0.02 M H 2 S0 4 , 0.1 M acetic acid (K = 
(H+)(Ac - )/(HAc) = 10" 5 ), and 0.1 M NH 4 OH (K = (NH+)(OH - )/ 
(NH4OH) = 1.8 x H)" 5 )? 

2. What is the pH of a solution containing 0.1 M acetic acid and 0.0 1 M, 0.1 M, or 
0.5 M sodium acetate? 

3. A solution initially contains 0.1 M NH4OH and 0.1 M NH 4 C1. Plot the pH 
change when acid and base are added to this solution. Compare this curve to 
the pH curve of a NH4OH solution titrated with HC1. Why are they different? 

4. What is the ionic strength of a solution containing 0.015 M Ca 2 " 1 ", 0.05 M Na + , 
0.030 MCI - , 0.02 M SO 2 . - , and 0.01 M HCO~? What are the activity coeffi- 
cients of the ions in this mixture? 

5. What is the aqueous concentration of H2CO3 plus CChaq in equilibrium with 
air? In equilibrium with a soil atmosphere of Pqo 2 = 0.1? What are the con- 
centrations of HCO^" and COj - in equilibrium with Pco 2 = 0. 1 at pH 6, 7, and 
8? 

6. Convert the following values to SI units: 

(a) lOmeq/lOOg 

(b) 2 mmho/cm 

(c) lOOkg/ha-yr 

(d) 150 lb/acre 

(e) 9 millimicrons 

(f) 7.4 A 

(g) 10 me/100 g 

7. Calculate the AG change during the change from feldspar to gibbsite, soluble 
silica, and K + (all components in their standard states). What is the further AG 
change if all the activities of soluble silica and K + are 10"" 4 ? 

8. The (A1)(P0 4 ) solubility product of variscite is about 10~" 21 . Calculate the AG 
of the reaction of soluble silica with variscite to form kaolinite at standard con- 
ditions and at Si(OH) 4aq = 10 -4 . 

9. Give the expression for the solubility product and complex stability constant of 
the hypothetical substance MA. Why do the two expressions differ? When is 
that assumption invalid? 

10. The tendency of minerals is to go to the lowest possible energy state. Why, then, 
are soils, which are metastable, formed? 

11. Describe the stepwise process of changes in water when an ion enters an aqueous 
solution. What are the two major characteristics of an ion that govern the ion's 
interaction with water? 



1 06 WATER AND SOLUTIONS 

12. What effect does one ion in solution have on another? How are these effects 
taken into account? 

13. What masses of which sails would be required to produce 1 L of the solution in 
Problem 4 (assume the salts are water-free)? 

14. Calculate (a) the activity of Fe 3+ in equilibrium with Fe(OH) 3 at pH 7 and 
(b) the activity of FeEDTA - in equilibrium with Fe(OH) 3 at pH 7 if 0.01 M 
H4EDTA was added to the solution. 

15. Calculate the pH in (a) a 0.005 M solution of HC1, and (b) a 0.005 M solution of 
acetic acid. Explain why the pH values are different. 

16. For a pesticide initially present at a concentration of 25 mg/kg of soil and having 
a half-life of approximately 15 days under the present field conditions, what 
concentration of the pesticide will remain after 1 10 days? 



4 



OXIDATION AND 
REDUCTION 



Oxidation and reduction dramatically change the behavior of the chemical elements. 
Oxidation is the loss or donation of electrons by an element; reduction is the gain 
or acceptance of electrons. Oxidation of one substance and reduction of another al- 
ways occur together — free electrons do not exist in chemical reactions. A substance 
can donate electrons only if another substance can accept them. The importance of 
oxidation-reduction (redox) reactions is that energy, the energy of life, is transferred 
by these electron transfers. Oxygen, carbon, nitrogen, and sulfur — and to a lesser ex- 
tent, iron and manganese — are the primary elements that carry out electron transfer, 
energy transfer, in the metabolism of living organisms and in soils. 

The hydrogen ion H + and the electron e~ have been called the two master vari- 
ables that govern chemistry. Together, the availability of H + and e~ determines the 
direction and rate of almost all organic and many inorganic reactions. The availabil- 
ity of H" 1 " and that of e~ are similar conceptually but different in reality. The H + 
availability, or potential, is related to its concentration in water and can be measured 
by the familiar pH glass electrode. The glass membrane around the electrode shields 
the electrode from other possible reactions. The measurement is so good that some 
people define pH as the H + potential, even though ions cannot be measured unequiv- 
ocally. The measured pH is probably very close to the H + potential, but we cannot 
know for certain. The pH electrode is the best example of ion-selective electrodes, 
where a selective membrane isolates the electrode from all substances but the desired 
substance in an aqueous solution. 

In contrast, the e~ availability and e~ potential are measurable under some con- 
ditions, but there is no concentration of free electrons that corresponds to the H + 
concentration. Someone calculated that the concentration of free electrons is about 
10 -45 M, or about 1 free electron per galaxy. Also in contrast to H + , the measure- 
ment of the e~ potential is qualitative in natural systems. Because the measuring 

107 



108 OXIDATION AND REDUCTION 

electrode is nonselective and has no shield analogous to the glass membrane of the 
pH electrode, many ions and molecules can donate and accept electrons from the 
measuring electrode. Each substance has its own degree of electron availability and 
hence its own potential at the electrode. The electrode's voltage is therefore a mixed 
potential, a weighted sum of the potentials of all the electron transfers at the electrode 
surface. The potential does not represent the e~ potential of any ion or molecule at 
the electrode. When only one substance is present in the system, the electron's po- 
tential may be measurable if the kinetics of electron transfer between the substance 
and the electrode surface are favorable. 

Virtually all biological reactions and many inorganic reactions in soils are redox 
reactions. Dioxygen O2 is the major and final electron acceptor {oxidizing agent) 
in nature and therefore buffers the e~ availability in aerobic systems (where O2 is 
available). Dioxygen diffuses through soil pores to plant roots, soil microbes, and 
inorganic substances from the atmosphere. Until the soil becomes quite wet and/or 
the oxygen demand is high, the oxygen diffusion rate is usually rapid enough to 
maintain adequate oxygen availability. Even if only the larger soil pores are open to 
the atmosphere, the oxygen supply can be sufficient because gas diffusion through 
the gas phase is 10 000 times faster than gas diffusion through water. 

If the diffusion path length from the soil surface through the soil solution is long, 
combined with a high oxygen demand from actively metabolizing roots and mi- 
crobes, oxygen may be lacking. Oxygen deprivation {anaerobic conditions) slows the 
rates of root metabolism and ion uptake, weakens plant resistance to soil pathogens, 
and increases the concentration of undesirable reduced ions in the soil solution. Oxy- 
gen dissolved in water in a flooded soil can supply oxygen for about 24 hours to 
plants. 

Much of the earth's soils are flooded or very wet for part or all of the year, subsoils 
have restricted water drainage and low oxygen concentrations, and the interior pores 
of soil aggregates in aerobic soils can have considerably lower oxygen concentrations 
than does the atmosphere. The ocean depths have restricted access to atmospheric 
oxygen. We think of aerobiosis as the ideal and common state of nature, but that is 
egocentrism. Most of the living earth has a limited oxygen supply. 

Carbon dioxide is formed when oxygen accepts electrons from carbon compounds 
during metabolism. The CO2 diffuses away in the same way as oxygen diffuses 
toward the reaction site. If diffusion rates are slow, as in flooded soils, the CO2 
and H2CO3 concentrations increase and they begin to buffer pH. The pH range of 
oxygen-deprived regions and poorly drained soils is therefore narrower than that of 
well-drained soils. 

Agricultural practices can change the soil's ability to supply oxygen. Irrigation, 
cultivation, introduced crops, lower plant densities, and the shorter growing season 
of agricultural crops change the soil's water content and therefore the change the 
pore space available for gas transfer between the root zone and the atmosphere. For 
example, large areas of the midwestern North America have tile drains to remove 
the water that accumulates during the crop season and reduces oxygen availability 
to roots. The cultivated plants are less dense and have a shorter growing season than 
native plants so water accumulates more. The native plant population transpires more 



SOIL OXIDATION-REDUCTION 1 09 

water and lessens water accumulation in the soil. Cultivation destroys the large soil 
pores through which gases and water move most rapidly. Cultivation also destroys 
the organic matter content, which maintains an open soil structure and improves 
permeability. 



4.1 SOIL OXIDATION-REDUCTION 

Redox reactions in the soil are mostly the result of a cycle started by photosynthesis. 
One part of the reaction is 

C0 2 + 4e - + 4H + = CH 2 + H 2 (4.1) 

where CH 2 represents a carbohydrate. Carbon in C0 2 accepts electrons and its 
oxidation state changes from the C 4+ in C0 2 to C° in carbohydrate (CH 2 0)„. 
Simultaneously, the 2 ~ in water gives up electrons as it oxidizes to 0° in 2 : 

2H 2 = 2 + 4e" + 4H+ (4.2) 

In these reactions 2 ~ is the electron donor, and C 4+ is the electron acceptor. Equa- 
tions 4. 1 and 4.2 are called half-reactions because they describe only half of the reac- 
tion. Although half-reactions appear to imply that free electrons exist, half-reactions 
imply only that the other half of the reaction is unspecified. The overall reaction of 
photosynthesis is the sum of the half-reactions: 

C0 2 + H 2 = CH 2 + 2 (4.3) 

Respiration (oxidation) in plants and animals and oxidation in soils complete the 
photosynthetic cycle by utilizing the energy stored in the carbohydrates and organic 
compounds derived from the carbohydrates, by disposing of organic wastes, and by 
producing the C0 2 needed for more photosynthesis by the reaction: 

CH 2 + H 2 = C0 2 + 4e~ + 4H + + energy (4.4) 

To obtain the energy and complete the reaction, organisms must find an electron 
acceptor to accept the electrons. If oxygen is available, the half- reaction of aerobic 
electron acceptance is the reverse of Eq. 4.2: 

2 + 4e _ + 4H + = 2H 2 (4.5) 

Equation 4.4 summarizes the many steps of the intricate Krebs or citric acid 
cycle that organisms utilize to obtain the energy in a useful form. Equation 4.5 
also oversimplifies the intricate mechanism of electron acceptance by oxygen in 
living organisms. 

A key to obtaining the energy in organic compounds, and thus to sustain life, is to 
obtain an electron acceptor. Higher plants and animals can utilize only 2 as an elec- 
tron acceptor, but microbes in soils and elsewhere can also utilize the oxidized states 



110 OXIDATION AND REDUCTION 

of nitrogen, sulfur, iron, manganese, and other elements as electron acceptors. These 
electron acceptors do not release all of the photosynthetic energy and retain some of 
the energy in the reduced states. The energy content of these reduced states makes 
them reactive and potential pollutants in our concept of a healthy, that is, aerobic, 
environment. Reacting further with O2 changes the partially oxidized compounds to 
higher, more benign, oxidation states. 

Redox reactions of C, N, and S compounds are catalyzed by enzymes. Catalysis 
is necessary because most elements exchange electrons reluctantly. Enzymes lower 
the activation energy of electron transfer and increase reaction rates enormously. The 
reluctance of C, N, and S compounds to reach equilibrium creates the metastability 
of carbon compounds and prevents you the reader and the paper of this page from 
immediately oxidizing to CO2. The irreversibility of electron transfer is a nuisance 
for physical chemists who like the simplicity of equilibrium, but is essential for life. 



4.2 ELECTRON DONORS 

The major electron donors in soils are the carbon compounds in living roots and 
microbes, in dead plant matter, and in soil organic matter (SOM). Table 4. 1 shows 
the approximate C, H, and O contents of the two largest plant components, cellulose 
and lignin, and of typical SOM. For simplicity, Table 4. 1 ignores the amounts of N, 
S, P, and other elements in these materials. Assuming that plant matter contains 1/3 
lignin and 2/3 cellulose, an empirical formula for "plant matter" is approximately 
C1.7H2.2O. Assuming further that all the carbon in this material oxidizes to C 4+ in 
CO2, the half-reaction of plant matter oxidation is 

C1.7H2.2O = 1.7C 4+ + H 2 + 0.2H+ + 7e _ (4.6) 

The reaction does not go to completion immediately. Some carbon remains as 
SOM — microbial biomass and partially metabolized by-products. 

The SOM in Table 4.1 is richer in carbon than is plant matter. SOM tends to 
contain more aromatic (cyclic and resonating carbon-carbon bond) compounds, and 
contains less oxygen, than the plant matter from which it is derived. Alternatively, 
because cellulose oxidizes faster than lignin, the aromatic groups may represent an 
accumulation of aromatic carbon from unreacted lignin. All of these materials even- 
tually oxidize in soils, but each succeeding oxidative step is much slower. The half- 



Table 4.1. Approximate C, H, and O composition of lignin, cellulose, and soil organic 
matter (nitrogen, sulfur, and other elements are ignored) 

C(%) H(%) 0(%) Empirical Formula 

Lignin 61-64 5-6 30 C2.sH2.9O 

Cellulose 44.5 6.2 49.3 C|. 2 H 2 

Soil organic matter 58 5 36 C2.2H2.2O 



ELECTRON ACCEPTORS 111 

reaction for the oxidation of soil organic matter is 

C2.2H2.2O = 2.2C 4+ + H 2 + 0.2H+ + 9e _ (4.7) 

Soil organic matter contains amino ( — NH2) and sulfhydryl ( — SH) groups, which 
also are electron donors. 

Inorganic electron donors in soils are generally in much smaller amounts and 
include sulfide S 2_ , sulfur S°, Fe 2+ , Mn 2+,3+ , and ammonia N 3_ . The reduced ox- 
idation states of the trace elements Cr, Cu, Mo, Hg, As, and Se are also electron 
donors in soils. 

4.3 ELECTRON ACCEPTORS 

The role of soil in the oxidation of reduced C compounds is to provide electron 
acceptors for plant roots and microbes. Oxygen is the strongest electron acceptor in 
nature and yields the most energy from oxidation (Eq. 4.5). Oxygen is also the only 
electron acceptor that plant roots can utilize. Oxygen is made available by diffusion 
through soil pores and by being dissolved in the soil solution. At soil temperatures, 
the dissolved O2 concentration is about 10 mg L _1 so that the O2 in air and water 
are about the same, on a volume basis. 

The O2 supply can be insufficient because soil pores are water-filled. The O2 
supply in unsaturated soils can also be less than the microbial and plant root demand 
due to a large supply of readily decomposable organic matter. Plant root demand for 
O2 is relatively constant while microbial demand fluctuates widely in response to 
organic inputs. High oxygen demand, relative to oxygen supply, also occurs in soils 
affected by leaks from natural gas pipes or used for organic waste disposal. Since 
oxygen diffusion from the surface is relatively slow, oxygen becomes deficient. 

Soil microorganisms, in contrast to higher plants and animals, can utilize other 
electron acceptors if O2 is unavailable. The prominent secondary electron acceptors 
in soils and their half-reactions are 

FeOOH + e~ + 3H+ = Fe 2 " 1 ' + 2H 2 (4.8) 

2MnOj .75 + 3e~ + 7H + = 2Mn 2+ + 3.5H 2 (4.9) 

where MnOi.75 signifies the complex Mn(III-IV) oxides in soils. 

SO 2- + 8e~ + 8H+ = S 2 ~ + 4H 2 (4. 1 0) 

NO3 + 8e~ + 9H + = NH 3 and amino acids + 3H 2 (4. 1 1) 

NO J + 2e~ + 2H 1 " = NO^" + H 2 (4. 12) 

2NOJ + lOe" + 12H+ = N 2 + 6H 2 (4. 13) 

2NO^ + 8e~ + 10H+ = N 2 + 5H 2 (4.14) 



112 OXIDATION AND REDUCTION 

The conditions that govern the endproducts of the nitrogen reactions are not yet well 
understood. The formation of N2O (nitrous oxide) is of interest because it is a long- 
lived "greenhouse" gas in the atmosphere and its concentration is increasing. Nitrous 
oxide is often released initially after nitrate fertilizers are added to soils. 

In addition to yielding less energy than O2, these secondary electron acceptors 
also yield products unfavorable to agriculture and aquaculture. The reduced oxida- 
tion states are more toxic than the oxidation states that are stable in the presence 
of02. Ammonia and nitrite, for example, are more toxic than nitrate; H2S is more 
toxic than sulfate. Reduction of Fe(III) and Mn(HI-IV) can cause phytotoxic Fe 2+ 
and Mn 2+ concentrations in rice paddies. Reduction of NO^~ to gaseous N2 and N2O 
is agriculturally undesirable because soil nitrogen is lost. 

If both oxygen and secondary electron acceptors are absent, microorganisms in 
soils and other systems can still extract some energy from photosynthetically pro- 
duced compounds by fermentation. Microbes can rearrange carbon compounds into 
more stable structures and release about 10% of the total energy in the initial com- 
pound. The products of fermentation include ethanol (C2H5OH), methane (CH4), 
peat, and CO2. In geologic time, further nonmicrobial reactions produce coal and 
petroleum. The fermentation products retain about 90% of the original energy and 
are useful fuels. 

Fermentation and reduction of secondary electron acceptors are temporary ex- 
pediencies for soil microbes. The resulting products are unstable in the presence 
of oxygen and eventually oxidize further when oxygen becomes available. SOM is a 
beneficial result of incomplete oxidation and fermentation. The SOM content reflects 
the difference of the rates of organic matter addition vs. oxidation rates. The rate of 
addition is essentially the rate of net photosynthesis. The oxidation rate is governed 
by temperature and by the rate of oxygen supply. 



4.4 REDOX REACTIONS 

The tendency of a substance to donate or accept electrons, and the measure of the 
electron's availability, is given by its electrode potential. In principle, electrode po- 
tentials can be measured directly by an electrode and a voltmeter. All chemical ele- 
ments can transfer electrons and thus change their oxidation states. Table 4.2 shows 
the standard electrode potentials of the half-reactions of several elements. The gen- 
eral reaction is 

Ox + e~ = Red (4.15) 

where Ox is an oxidized state of the element and Red is a reduced state. 

High electrode potentials mean that Ox is available and readily accepts electrons. 
The halogen gases, for example, have high electrode potentials and thus are strong 
oxidizing agents; they want to accept electrons and be reduced. Low electrode po- 
tentials mean that Red is available and readily donates electrons. The alkali metals 
are strong electron donors, are strong reducing agents, and are avid electron donors. 



REDOX REACTIONS 113 



Table 4.2. Electrode potentials (reduction potentials) of 
selected half-reactions at 25° C. (The dashed lines show the 
limits of electrode potential in aqueous systems). 



Reaction 




Eh° (V) 


F 2 + 2e~ = 2F- 




+2.87 


Cl 2 + 2e" = 2Cr 




1.36 


NO^ -I- 6H+ + 5e- = ^N 2 + 3H 2 


1.26 


2 + H+ + 4e- = 2H 2 




1.23 


NO~ + 2H+ + 4e' = NO~ 


+ H 2 


0.85 


Fe 3 + + e~ = Fe 2+ 




0.77 


S0 4 + 10H+ + 8e' = H 2 S + 4H 2 


0.3 1 


C0 2 + 4H+ + 4e~ =C + 2H 2 


0.2/ 


N 2 + 6H+ + 6e - = 2NH 3 




0.09 


2H+ + 2e- = H 2 







Fe 2+ + 2e _ = Fe 




-0.44 


Zn 2+ + 2e~ = Zn 




-0.76 


Al 3+ + 3e- = Al 




-1.66 


Mg 2+ + 2e~ = Mg 




-2.37 


Na+ +e~ = Na 




-2.71 


Ca 2 + + 2e" = Ca 




-2.87 


K+ + e - = K 




-2.92 



The range of electrode potentials possible in soils is limited by the stability of 
water with respect to oxidation and reduction. High electrode potentials can oxidize 
water to 2 . Low electrode potentials can reduce water to H 2 . If a solution contains 
an oxidizing agent such as Cl 2 with an electrode potential greater than that of the 
H 2 0-0 2 couple, or half-reaction, the oxidizing agent can oxidize water to 2 : 

i0 2 + 2e"+2H + = H 2 (4.16) 

The oxidation of water to 2 prevents the electrode potential from rising above the 
electrode potential of Eq. 4.6. Strong oxidizing agents (high electrode potential) such 
as hypochlorite (CIO - ) and Cl 2 are unstable and decompose in soils and water by 
reducing to CP as they oxidize water to 2 . Although oxidizing agents stronger than 
2 should not in principle be formed in an aqueous system, N 2 is an apparent ex- 
ception. Nitrous oxide has a high electrode potential, but nitrogen electron transfers 
are sluggish and irreversible. Not only is N 2 formed in soils, it is somewhat stable 
in soils and is stable for many years in the atmosphere. 

Strong reducing agents, in contrast, can reduce water to H 2 : 

2H + +2e~=H 2 (4.17) 



114 OXIDATION AND REDUCTION 

The H + ion is reduced rather than H2O, but H + is always present in aqueous solu- 
tions because water dissociates. The H + -H2 couple (Eq. 4.17) is the lower limit of 
electrode potential in aqueous systems and soils. 

The stability of water with respect to oxidation to O2, and reduction to H2, lim- 
its the range of electrode potentials and the oxidation states possible in any system 
containing water. These limits are the dashed lines in Table 4.2. Dissolved F2 and 
Cb gases are unstable because of their high electrode potentials. The reduced states 
F - and Cl~ are the stable states in water because they have accepted an electron and 
thereby discharged their oxidizing power. Metals are likewise unstable in water and 
soils because of their tendency to oxidize, to donate electrons. Their oxidized states 
(Al 3+ , Ca 2+ , K + , etc.) are stable in water. Table 4.2 implies correctly that the com- 
mon metals are unstable and will corrode. Exceptions such as aluminium and zinc 
metals are metastable; an oxide layer that forms initially on their surfaces inhibits 
further oxidation. Soils and seawater catalyze the breakdown of these protective lay- 
ers and speed up their corrosion (oxidation). Iron and steel do not form this protective 
layer. 

For the redox couples between the dashed lines in Table 4.2, both states are stable 
in water and soil, depending on the electron availability. Under reducing conditions 
Fe 2+ , sulfide, and ammonia are stable. If 2 is available, Fe(lll), SO 2 .", and NO3 
are stable. 

Soil microorganisms utilize the strongest electron acceptors available, in or- 
der to obtain the maximum energy from their food substrate. If the O2 supply 
is insufficient, the next strongest electron acceptor available in soils is nitrate. The 
manganese(IV,HI-II) redox couple may be stronger, but Mn(IV,III) oxides are solids, 
which cannot diffuse to the microbes, so their availability is low. Nitrate reduces to 
amino acids, N2 or N2O. After oxygen and nitrate have been exhausted, Fe(III) and 
Mn(IVJH) hydroxyoxides can be reduced, and Fe 2+ and Mn 2+ concentrations in 
the soil solution increase. If the rate of electron acceptance by these acceptors is 
less than the availability of food sources, even stronger reducing conditions result. 
Microbes can reduce sulfate to sulfur or sulfide, ferment organic matter to CO2 and 
methane or peat, and in extreme cases reduce water to H2. 

The stepwise order of reduction in Table 4.3 is idealized. The rate of electron 
availability is usually much faster than the rate at which electron acceptors can dis- 
solve and diffuse to the soil microbes. The reactions then overlap and several proceed 
simultaneously. 

Under slower conditions in the laboratory, the order of utilization of electron ac- 
ceptors follows the order of electron potentials at pH 7, as listed in the second column 
of Table 4.3. The standard electrode potentials of Table 4.2 are at pH 0. Most elec- 
trode potentials change similarly with pH, however, so the order of standard electrode 
potentials and of utilization of electron acceptors are similar. Appendix 4. 1 describes 
how the pH dependence of electrode potentials can be calculated. 

The third column of Table 4.3 lists the measured ranges of potentials over which 
each reaction occurs in soils. The potentials are measured with a platinum electrode 
whose potential responds roughly to the electron availability. Reasons for the dif- 
ference between electrode and redox potentials are discussed in Appendix 4.2. In 



FLOODED SOILS 115 

Table 4.3. Order of utilization of principal electron acceptors in soils, equilibrium 
potentials of these half-reactions at pH 7, and measured potentials of these reactions in 
soils 







Measured Redox 






Potential in Soils 


Reaction 


Eh at pH 7 (V) 


(V) 


O2 disappearance 






\ 2 + 2e~ + 2H+ = H 2 


0.82 


0.6 to 0.4 


NOJ' disappearance 






NO^" + 2<T + 2H+ = N07 + H 2 


0.54 


0.5 to 0.2 


Mn 2+ formation 






Mn0 2 + 2e~ + 4H+ = Mn 2+ + 2H 2 


0.4 


0.4 to 0.2 


Fe 2+ formation 






FeOOH + e~ + 3H+ = Fe 2+ + 2H 2 


0.17 


0.3 to 0.1 


HS~ formation 






SO^ + 9H+ + 6e~ = IIS - + 4H 2 


-0.16 


to -0.1 5 


H 2 formation 






H++e" = iH 2 


-0.41 


-0.15 to -0.22 


CH4 formation (example of fermentation) 






(CH 2 0)„ = «/2 C0 2 + n/2 CH 4 


— 


-0.15 to -0.22 



addition, part of the reason for the range in Table 4.3 is that the potentials were 
measured at soil pH values other than 7, and redox potentials are pH dependent. 

Redox conditions in soils vary widely over short distances because 2 must dif- 
fuse through pores of various sizes and water-filled pores. In aerobic soils the interior 
of soil aggregates may be partially anaerobic. The change from oxygen sufficiency 
to deficiency can occur within a few millimeters. In wet soils, only the largest pores 
are open to gas diffusion from the atmosphere. 

Redox conditions in saturated and flooded soils can be more homogenous and 
redox measurements with the platinum electrode tend to be more reliable. Some 
water-loving plants such as rice, however, can conduct oxygen through the stern to 
the root. The soil immediately surrounding such roots is oxidized compared to the 
rest of the flooded soil. Convection currents in overlying water also bring oxygen 
downward, so that submerged sediments are often topped by a thin oxidized layer. 



4.5 FLOODED SOILS 

Ponnamperuma (1972) reviewed the chemistry of flooded soils. This research has 
understandably concentrated on the soil conditions of paddy rice agriculture. Only 
some generalizations are mentioned here. The behavior of C, N, S, Fe, and Mn gen- 
erally follows that shown in Table 4.3. When rice paddies are drained before harvest, 
redox potentials rise, Fe 2 ' 1 ' and Mn 2+ concentrations decrease, and C, N, and S oxi- 
dize. When the soils are flooded again, the reactions reverse. 



116 OXIDATION AND REDUCTION 

Phosphate apparently precipitates as Fe(III) and Al phosphates during the dry part 
of the rice culture cycle. Under subsequent reducing conditions, the Fe(III) phosphate 
is reduced to more soluble Fe(II) phosphate. This reduction can account for the rather 
high availability of phosphate for centuries in paddy soils. The Fe(III) phosphate 
may be slightly more stable than Al phosphate so the Al phosphate that precipitates 
initially slowly transforms to Fe(III) phosphate. Similar aerobic soils often supply 
inadequate phosphate to plants because Fe(III) phosphate remains insoluble. 

A second noteworthy flooded soil is acid sulfate soil. Sediments along tropical 
and subtropical coastlines and river deltas may contain significant quantities of Fe(II) 
sulfides. When drained, these sulfides oxidize to H2SO4 and the acidic Fe 3+ ion. The 
soil acidity can increase to pH 2. Such conditions are highly phytotoxic and can be 
remedied under aerobic conditions only by extensive leaching and lime applications. 
If resubmerged, acid sulfate soils revert rapidly to near neutrality as the Fe(III) and 
sulfate are reduced back to Fe(II) sulfides. 

A third example of flooded soils is the extensive areas of soils rich in organic 
matter, peal and muck soils, or Histosols. The slow rate of organic matter oxidation 
is due to slow O2 diffusion through stagnant water; to low concentrations of mineral 
nutrients; and, in the most extensive areas of Histosols in Canada and Siberia, to 
low temperatures. Peat soils are to some degree self-perpetuating. They have a high 
water-holding capacity, which allows them to spread up slight slopes above the water 
table. In warm seasons the high specific heat of water slows the warming rate of 
the peat. During occasional dry seasons the high thermal resistivity of peat slows 
warming below the immediate surface. 



APPENDIX 4.1 ELECTROCHEMISTRY 

Electrode, electron, reversible, and equilibrium potentials and electron activity are 
closely related terms for the equilibrium potential of the electron, where potential 
means availability and driving force. Equilibrium implies that all electron transfers 
are reversible, that a small change of electron potential will bring about a correspond- 
ing electron transfer. Table 4.2 lists such reversible potentials for half-reactions of 
some common ions. Reversible election transfers, however, are rare. Irreversible re- 
actions (Appendix 4.3), in which the electron transfer is hindered by an activation 
energy barrier, are the norm. This barrier can be overcome by extra energy or over- 
voltage. Enzyme catalysis and soil surface catalysis can reduce, but not completely 
remove, this activation energy barrier. As a result of irreversibility, redox reactions 
in soils respond sluggishly to changes in electron potential. The irreversibility is rel- 
atively low for iron reactions; carbon and nitrogen electron transfers are exceedingly 
irreversible. 

Oxidation-reduction equilibrium also implies that the electrode potentials of all 
redox couples in the system are equal. Because of irreversibility, this condition is rare 
in mixtures of redox couples, especially in mixtures containing organic and nitrogen 
compounds such as the soil solution. 



ELECTROCHEMISTRY 117 

Despite these limitations, equilibrium is a convenient starting point from which 
to study redox reactions. A generalized redox half-reaction is 

Ox + n&~ + mU + = Red + -HbO (4. 1 8) 

2 " 

where Ox is the oxidized species and Red is the reduced species of the redox couple. 
The equilibrium equation for this reaction is 

Red 

K = ■ — (4.19) 

Ox(e-)"(H+)'« 

where (e~) has been called election availability, electrode potential, and electron 
activity. The electron activity has no relation to ion activities in solution because the 
concentration of free electrons in solution is vanishingly small. 

Electron (or electrode) potential is given the symbol Eh, while electron activity 
is associated with pe. Equation 4.19 can be transformed into the Nernst equation: 

„ RT , Red 0.059 Red 0.059 

Eh - Eh° = — log — = log pH (4.20) 

nF fe (Ox)(H+)'" n & Ox n/m 

where Eh° is the standard electrode potential, R is the gas constant, T is absolute 
temperature, F is the Faraday constant, and 0.059 is the quotient of the constants 
at 25° C. The standard potential must be defined in terms of an arbitrary reference 
state because, like free energy, the absolute potential cannot be defined. The Eh° is 
related to the standard Gibbs free energy by 

G° = -n F Eh° = -96.48/? Eh° (4.2 1 ) 

where G° is in joules. 

The electron potential pe is an alternative expression of electron availability. From 
Eq.4.19, 

pe + pH = log K - log(Red) + log(Ox) (4.22) 

when n = m. Adding pe and pH is advantageous because the sum can be used as 
one axis of a two-dimensional graph, while the activities of the reduced and oxidized 
species are plotted on the other axis. Adding pe + pH also makes arithmetic sense 
because the numerical range of pe and pH are similar. Each has approximately equal 
weight in the sum, in accord with their joint importance at equilibrium. Adding pe + 
pH, however, implies that both are at equilibrium, and that is rarely the case for pe. 
Nonetheless, pe + pH diagrams can summarize a great deal of thermodynamic data 
and provide a picture of the behavior of chemical elements in nature. 
At 25°C, 

pe = 0.059E/J (4.23) 



APPENDIX 4.2 Eh AND pe 

The range of electrode potentials in systems containing water is limited by the stabil- 
ity of water with respect to oxidation and reduction (Eqs. 4.16 and 4.17). Substituting 
into the Nenist equation (Eq. 4.20) yields 

Eh = Eh° - ^^1 i g _L _ 0.059 pH (4.24) 

4 "o 2 

for the upper limit of oxidizing conditions in soils and water, and for the lower limit, 

„ 0.059 1 

Eh ~ Eh° log 0.059 pH (4.25) 

2 Ph 2 

where Pq 2 and Ph 2 are Ihc partial pressures of oxygen and hydrogen gases. Gases 
are essentially ideal at low pressures so the partial pressure and activity are equal. 
The Eh is dependent on the pH and the gas concentration but changes only 15 mV 
per tenfold change of Po 2 and 30 mV per tenfold change of Ph 2 - The Eh limits of 
water, and therefore of soil and biological systems, are plotted in an Eh-pH diagram 
(Fig. 4.1a) at unit partial pressures of O2 and H2. Water is stable between the lines; 
dioxygen is stable at oxidizing conditions and electrode potentials above the upper 
line. Dihydrogen is stable at potentials below the lower line. 
Equation 4. 19 can be transformed to 

(H2O) 1 / 2 

X = — , ttj (4-26) 

(H+)(e-)(PoO ,/4 



and 



pe + pH - 20.78 + \ log Pq 2 (4.27) 



The pe +■ pH values range from 20.78 at highly oxidizing conditions at pC>2 = 1 and 
pH 0, to pe + pH = at strongly reducing conditions when pH2 = 1 (Fig. 4.1b). 

Carbon, nitrogen, sulfur, and iron are elements that lend themselves to redox con- 
siderations. Interested readers should consult Pourbaix (1966) and Garrels and Christ 
(1965) for Eh-pH diagrams for other elements, or Lindsay for pe 4- pH plots. All of 
these diagrams assume equilibrium and that Eh can be measured accurately. Both as- 
sumptions are questionable, but some success has been obtained measuring changes 
in Fe chemistry in soils. Carbon Eh-pH and pe + pH diagrams are not very inter- 
esting. The stability of CO2 totally dominates the diagrams. Carbon compounds are 
stable only in a narrow strip near the H + -H2 boundary, that is, they are stable only 
under strong reducing conditions. 

A4.2.1 Nitrogen 

Eh-pH and pe + pH diagrams can qualitatively describe nitrogen chemistry at the 
earth's surface, including simple interactions with biological systems. The chemistry 



E/iANDpe 119 



1.5 r 



1.0 



0.5 




(a) ^~ 
w w o 



-0.5 



-1.0 




_i — i i_i — i — i — i — ■_ 



7 
pH 



14 



(b) 



'm 2 


= 1 

Reducing 
conditions 




P o 2 

Oxidizing 
conditions 


= 1 


H 2 

I 


1 1 1 


H 2 

1 1 1 


Jill 


2 

I 



10 

pe + pH 



20 



FIGURE 4.1. The range of oxidizing and reducing conditions as shown in (a) Eh-pH and (b) pe 
and pH diagrams. The range is defined by the breakdown o< water to H2 or O2. 



of nitrogen is usually discussed in terms of the microorganisms and the reaction 
mechanisms that they carry out. Microbes and enzymes are only catalysts, however, 
and perform only those redox reactions that electron availability permits. Eh and pe 
describe that electron availability. 

Nitrogen has many oxidation states that might be stable within the redox stability 
range of water. By comparing the electrode potentials of all possible redox couples, 



120 



OXIDATION AND REDUCTION 



the stable oxidation states can be sorted out. For example, nitrate is reduced to ni- 
trite (NOj ) at Eh° = 0.95 volts (V), and nitrite reduces to N 2 at Eh° = 1.25 V. 
Those potentials mean that if a reducing agent is added to a nitrate solution, all of 
the nitrate will reduce to N2 before the electrode potential increases enough to re- 
duce nitrate to nitrite. Nitrite ions are therefore unstable and should spontaneously 
decompose to N 2 . Nitrite solutions are indeed unstable but decompose only slowly, 
because nitrogen redox equations are invariably irreversible. 

The stable oxidation states of nitrogen within the stability limits of water turn out 
to be nitrate, N 2 , and ammonia. Their oxidation-reduction relationships are shown 
in Fig. 4.2. The stability region is dominated by N 2 . Nitrate is stable only at strongly 
oxidizing conditions and pH > 3. Ammonia is stable only under reducing condi- 
tions. Although not shown, the stability regions of NO^ and NH3 are concentration 
dependent, increasing slightly as the their solution concentrations decrease. 

Figure 4.2 also shows the stability region of the amino acid alanine (CH3 — 
CHNH 2 — COOH), whose stability is typical of other amino acids. Nitrogen in ala- 
nine is more stable to oxidation than is ammonia. This accounts for the rarity of 
ammonia in nature compared to amino nitrogen. Ammonia is liberated when the car- 
bon required for amino compounds disappears, such as in the composting of manure 
and in wastewater treatment. The alanine-N 2 boundary is shown as a dashed line 
because it is not a true equilibrium. Alanine is unstable with respect to oxidation to 
C0 2 . The alanine-N 2 boundary is a steady state and is maintained by the continuous 
production of amino acid precursors by photosynthesis. 

A protein-N 2 boundary unfortunately cannot be calculated, because the energies 
of formation of proteins are not yet known. Proteins are presumably more stable than 
amino acids, because free amino acids such as alanine are rare in nature, while pro- 
teins are common. The protein-N 2 boundary would represent the highest potential 




FIGURE 4.2. Eh-pH diagram of the N-H20-02-amino acid system. 



Eh AND pe 



121 



N 2 (g) 




8 10 12 14 16 18 20 22 
pe * pH 

FIGURE 4.3. Distribution of soluble and gaseous nitrogen species. Total soluble nitrogen is 
assumed to be 10~ 3 M. (Adapted from Lindsay (1979).) 



(the strongest oxidizing condition), at which Nt could be converted to protein by soil 
microorganisms or other means during nitrogen fixation. 

Figure 4.3 shows some pe + pH relations of nitrogen. This diagram has the ad- 
vantage of being able to plot the concentration dependence of the redox equilibria. 
The vertical axis is the aqueous activity of dissolved ions or the partial pressure of 
gases. The concentration of soluble nitrogen is assumed to total 10~ 3 M and the N2 
partial pressure is assumed to be that of the atmosphere, 0.78. Nitrite and nitrous 
oxide can be present in only exceedingly low concentrations at equilibrium. Higher 
concentrations of NO J and N2O (and NO and the NO^-N20 4 pair) are unstable; 
they will tend to degrade to NH3, N2, or NOJ . 

The Eh-pH diagram shows only the "stable" nitrogen species over the range of 
electron potentials in aqueous systems. The pe + pH diagram shows that the "stable" 
species are actually only the most prominent ones. The "unstable" species are also 
present but in much lower concentrations. 

The overall distribution of nitrogen compounds roughly follows the distribution 
suggested by Figs. 4.2 and 4.3. Most of the world's nitrogen exists as N2. Some 
also exists as amino nitrogen in reduced carbon compounds, living organisms, and 
dead organic matter. A very small fraction exists as nitrates. If equilibrium existed, 
the nitrate fraction would be much larger because of oxidation of organic N, and 
oxidation of atmospheric N2 to NO^": 



N 2 + 2.502 + H 2 = 2HN0 3 



(4.28) 



Almost all the O2 in the atmosphere would be consumed. The remaining atmosphere 
would contain perhaps 99% N2 and 1% 2 , and the oceans, rain, and "freshwa- 
ters" would be dilute HNO3. The reaction is fortunately hindered by the general 



122 OXIDATION AND REDUCTION 

irreversibility of nitrogen reactions. Reaction 4.28 proceeds in the atmosphere to a 
limited extent when lightning provides sufficient activation energy. 

Nitrogen reactions are generally highly irreversible, and enzymatic catalysis is 
necessary for nitrogen conversions in soils. Redox irreversibility is unfortunate from 
the standpoint of studying redox reactions, but is absolutely necessary for life. Re- 
versibility would bring with it equilibrium, and all organisms would be transformed 
into C0 2 , N 2 , NO^~, and H 2 0. 

Irreversibility means that more energy is required to carry out, and less energy 
is derived from, reactions. While natural reactions are remarkably conservative of 
chemical energy, energy optimization is not the only criterion that determines which 
reactions occur. Unstable compounds, such as nitrite and nitrous oxide (N2O), for 
example, would not be produced if soil microorganisms were concerned only with 
optimal utilization of available chemical energy. Spring thaw can bring about tem- 
porary nitrite accumulation in soils, apparently because microbial nitrite reducers 
respond slower to increased temperature than do the microbes that reduce nitrate to 
nitrite. In addition, nitrate fertilization or wetting of dry soil initially stimulates N2O 
production, so that up to 25 to 50% of the nitrogen lost by denitrification can be lost 
as N2O. Both N2O and nitrite production appear to be temporary maladjustments 
during a flurry of microbial activity after a sudden environmental change. 



A4.2.2 Sulfur 

Figures 4.4 and 4.5 show the equilibrium relations of the sulfur oxidation states — 
sulfate, elemental sulfur, and sulfide — which are stable within the stability limits of 
water. The stable oxidation states are similar to those of nitrogen. Elemental sulfur, 
however, is much less stable than is elemental nitrogen and is stable only under acid 
conditions. 

Figure 4.4 shows the sulfur stability regions for unit activity of sulfate and sulfide. 
The region of sulfur stability diminishes with decreasing sulfate and sulfide concen- 
trations. Sulfur in the amino acid cysteine (HS — CH2 — CHNH2 — COOH) is more 
stable against oxidation than is the nitrogen in alanine. The cysteine sulfate bound- 
ary is shown as a dashed line because it is unstable at Eh° = 0.3 1 V with respect 
to CO2. Like alanine, the stability of cysteine is a steady state dependent on continu- 
ing production of organic compounds. The stability region of cysteine increases with 
increasing concentrations of sulfate compounds. 

Figure 4.5 is a pe + pH diagram of sulfur at pH 7 and 10 -3 M soluble sulfur 
species. Only SO^ - and HS"" are stable, with the HS~ stable only under strongly re- 
ducing conditions. The soluble sulfur concentration is too low and the pH is too high 
to permit elemental sulfur stability. Figure 4.5 ignores organic sulfur compounds. 

Sulfur redox reactions seem to be more reversible than those of nitrogen. Inter- 
mediate compounds in the reaction series from sulfate to sulfur or sulfide, and vice 
versa, do not appear in soils. Sulfur also differs from nitrogen in that little sulfur 
volatilizes from soils. Although H2S is a gas, apparently any H2S formed in soils re- 
acts rapidly with Fe and other transition metal oxides to form sulfides. Some organic 



EfiANDpe 123 




FIGURE 4.4. Eh-pH diagram of the S-H 2 0-0 2 -amino acid system. 



Or 



-20 



B -40 

t 
o 

Q. 



-60 - 



-80 - 



% -TOO 



O 
O 



-120 



-140 - 



--^ 


y, 


so/ 




x^^c 




/ 


/ 




^ SO^g) 


x^. 


^>^ 


- 


!/> 






- 






\ v' - 


- 




1 1 1 


i i\ — Aj 



8 12 16 

pe ♦ pH 



20 



24 



FIGURE 4.5. Distribution of solid, gaseous, and aqueous sulfur species at pH 7 and 10 3 M 
total soluble sulfur. (Adapted from Lindsay (1979).) 



124 



OXIDATION AND REDUCTION 



gases containing — SH groups are occasionally liberated during decomposition of 
fresh organic matter, but the amounts are small. 



A4.2.3 Iron 

Electron exchange of iron(II-III) tends to be more reversible than is electron ex- 
change between nitrogen, sulfur, or carbon states. Iron redox reactions occur in soils 
without enzymatic catalysis. The Fe(II) minerals in parent material rocks oxidize 
spontaneously, though slowly, in aerobic soils. The electron availability for subse- 
quent Fe redox reactions in soils is determined by microbial oxidation of carbon 
compounds. The reduction of Fe(III) in acid solutions is 



.3+ 



Fe Ji " + e" = Fe 



-r^2+ 



(4.29) 



The concentration of Fe(III) at soil pH levels is very low because of the insolubility 
of Fe(III) hydroxyoxides. As a result Eq. 4.29 consumes a negligible number of elec- 
trons. The major reaction by which Fe(III) accepts electrons in soils is the reduction 
of solid-phase Fe(III) hydroxyoxide: 



FeOOH + e" + 3H+ = Fe 2+ + 2H 2 



(4.30) 



Substituting Eq. 4.30 into the Nernst equation (4.20) yields 



Eh = Eh a - 0.059 



log(Fe 2+ )+3pH] 



(4.31) 




FIGURE 4.6. The Eh-pH diagram of various iron ions and compounds. 



Equations 4.31 and the Nernst form of Eq. 4.29 are two of the boundaries in the 
Eh-pH diagram for Fe (Fig. 4.6). The diagram shows that FeOOH dissolves to Fe 2+ 
under reduced and moderately acidic conditions. The Fe 3+ ion predominates un- 
der strongly acidic and oxidizing conditions. Fe 2+ and FeOOH are the predominant 
states in typical well-aerated soils. Goethite (FeOOH) is less stable at equilibrium 
than hematite (Fe203) in many free-energy tabulations, but goethite seems to govern 
the solution chemistry of Fe in soils. 

The electrode potential of Eq. 4.31 depends on the Fe 2+ concentration. The lines 
shown in Fig. 4.6 are for 10 -6 activity of soluble Fe. Higher activities shift the 
equilibrium boundaries to the left. Under strongly reducing conditions FeOOH is 
unstable with respect to Fe(OH) 2 , magnetite (Fe304), siderite (FeC03), and pyrite 
(FeS2). The region of FeC03 stability increases with increasing Pco 2 - The region 
shown corresponds to Pqq 2 = 0.01, which is not unreasonable for moist field soils. 



APPENDIX 4.3 REDOX POTENTIAL MEASUREMENTS 

Applying models of equilibrium oxidation-reduction, such as Figs. 4.2, 4.4, and 4.6, 
quantitatively to soils requires that the electrode potential be known. From the elec- 
trode potential one could then calculate the soil solution concentrations of Fe 2+ , 
Mn 2+ , and NO^ and the sulfate/sulfide ratio from Eq. 4.20. Ideally, the potential 
of an inert electrode in the system should equal the electrode potential, because the 
electrode should take on a potential corresponding to the electron availability. This 
measurement is called the redox potential. 

Redox potential measurements are even simpler than pH measurements. A plat- 
inum electrode with its necessary reference electrode is inserted into a soil or suspen- 
sion and the potential is measured with a sensitive potentiometer, such as a pH meter. 
Platinum is the preferred inert electrode material because of its greater response to 
change in redox conditions. 

Table 4.3 shows, however, that redox potentials often differ greatly from electrode 
potentials. Ion activities are only qualitatively related to redox potentials, except in 
rare circumstances. One reason is that the Nernst equation applies only to equilib- 
rium. Redox reactions in soils are nonequilibria, though in some cases for highly 
reduced soils, a steady state may be reached approximating equilibrium. Then only 
a few redox couples in the soil affect the platinum electrode and the result may ap- 
proach a pseudo-equilibrium. 

Second, equilibrium implies that the electrode potentials of all redox couples in 
the system are equal. Electrons would exchange between all redox couples until the 
potentials of all the available electrons are equal. In aerobic soils, this would mean 
that the potentials of all redox couples would have to equal the potential of the O2- 
H2O couple because it would be a major factor. Measuring the redox potential would 
be unnecessary. 

Third, redox couples do not necessarily transfer electrons equally or reversibly 
with platinum or other inert electrodes. Election transfers between ions or molecules 
and electrodes are usually irreversible even in pure solutions. Reversibility is unlikely 



126 



OXIDATION AND REDUCTION 



to be greater in a mixed system such as soils. The inert electrode responds more 
readily to reversible redox couples than to irreversible couples. The Fe(II-IIl) and 
H + -H2 couples are relatively reversible at platinum surfaces, so, if present, they 
strongly influence the potential of the electrode. 

Fourth, the relative effect of a redox couple on the redox potential increases with 
the couple's concentration. A high concentration affects the potential more because 
the likelihood of electron transfer with the electrode is greater. The redox potential 
is a measure both of the amount of electron transfer and of the electrode potential. In 
aerobic soils the O2-H2O couple is influential, despite its irreversibility, because the 
O2 concentration is high. 

The potential of the platinum electrode in a mixture of redox couples is a poorly 
defined, weighted average of the potentials of all the redox couples present. The 
contribution of each couple to the average potential is an unknown function of its 
concentration, irreversibility, and equilibrium electrode potential. The potential of a 
nonequilibrium mixture of redox couples is not the potential of any single couple 
and is a mixed potential. 

Under aerobic conditions the redox potential deviates widely from the potentials 
of soil redox couples. In anaerobic soils, redox potentials may be more quantita- 
tively related to ion activities. The Fe 2+ and perhaps Mn 2+ concentrations are high 
and tend to dominate the redox potential. The range of redox potentials that have 
been measured in soils is shown in Fig. 4.7. The envelope around those data was 
considered by the investigators to be the extreme limits of likely redox potentials 
and pH values in soils and natural waters. Redox potentials can closely approach the 
H + -H2 potential, because it is nearly reversible at the platinum electrode. 




10 12 



FIGURE 4.7. The range of redox potentials and pH values in soils. (From L. G. M. Bass Becking, 
I. R. Kaplan, and D. Moore. 1960. J. Geol. 68:243.) 



QUESTIONS AND PROBLEMS 1 27 

The redox potential is pH-sensitive and should be measured in conjunction with 
pH. To convert redox potentials to a common pH, the conversion factor of —59 mV 
per unit pH is usually employed, although the change of redox potential with pH 
can be as great as —200 mV per unit pH during rapid redox changes. The value of 
—59 mV per unit pH is probably reasonable for most measurements. Reporting both 
the measured redox potential and the pH avoids the conversion problem. Because 
redox potentials are most often mixed potentials, measurements move precise than 
±10 mV have little significance 

The redox potential, like the pH, is an intensity measurement. In theory, it mea- 
sures the availability rather than the quantity of electrons. Because the redox poten- 
tial is governed by the potential rate of electron donation versus the rate of electron 
acceptance, however, it is also a crude estimate of the number of electron donors (re- 
ducing agents) present. The qualitative nature of redox measurements has discour- 
aged their use. Redox measurements, however, are used frequently in wastewater 
treatment to indicate the extent of treatment necessary. 



BIBLIOGRAPHY 

Bonn, H. L. 1970. Redox potentials. Soil Sci. 112:39-45. 

Garrels, R. M., and C. L. Christ. 1965. Minerals, Solutions, and Equilibria. Harper, New York. 

Lindsay, W. 1979. Chemical Equilibria in Soils. Wiley, New York. 

Ponnamperuma, F. N. 1972. The chemistry of submerged soils. In Advances in Agronomy, 
Vol. 24. (N. C. Brady, ed.). Academic, New York, pp. 29-96. 

Pourbaix, M. 1966. Atlas of Electrochemical Equilibrium in Aqueous Solutions (English trans- 
lation by J. A. Franklin). Pergamon, Oxford, UK. 



QUESTIONS AND PROBLEMS 

1. Why do redox potentials measured in soils differ from electrode potentials? 

2. What electrochemical properties for redox measurements would the ideal elec- 
trode have? 

3. From the elemental analyses in Table 4. 1 , derive the empirical formula for SOM. 

4. Write the balanced reactions for the oxidation of plant matter (Eq. 4.6) and SOM 
(Eq. 4.7) by oxygen. 

5. Write balanced half-reactions and Nernst equations for the reduction NO^" to N2 
and for N2 toNH3. 

6. What would you expect the redox potential to be in a soil of pH 7 at field mois- 
ture capacity? If the soil were at equilibrium, what should tbe electron potential 
be? Calculate the exact potential from Eq. 4.21 



1 28 OXIDATION AND REDUCTION 

7. From the hydrolysis reactions of iron given in Chapter 3 and from the Nernst 
equation (Eq. 4.20), derive the equation for the Fe(II-III) potential as a function 
of H + activity from pH to pH 14. 

8. Assuming equilibrium, what is the Eh value of the O2-H2O water couple at 
Pq 2 = 0.21 (normal air), 0.20 (typical soil air), and 0.01 (perhaps typical of the 
air in an anaerobic soil aggregate)? What ranges of redox potentials might a plat- 
inum electrode measure in these three situations? Why do the redox potentials 
differ from the electrode potential? 

9. Describe the likely chemical changes in a soil as it becomes increasingly anaer- 
obic. 

10. If oxidizing agents with electrode potentials greater than the H2O/O2 couple 
should not persist in aqueous solutions, why are solutions KM11O4 and NaOCl, 
which are common in the laboratory and home, rather stable? 

11. What redox reactions tend to "poise" the redox potentials of most soils in the 
"normal" region of Fig. 4.7? 



5 



INORGANIC SOLID PHASE 



Soil consists of inorganic and organic solids, the soil solution, soil air, and living 
organisms. Soil chemistry is primarily concerned with the solid and liquid phases 
and their interactions. This chapter deals with the inorganic solid phase of soils, 
primarily the clay fraction, and its important properties. The clay fraction (<2 /Am in 
size) carries out most of soil chemistry. The larger sand (50-2000 /xm) and silt (2-50 
fim) fractions are much less chemically active and are composed largely of quartz. 
(SiO?), which is rather inert. 

Inorganic soil particles range in size from colloidal particles (<2 jxm) to gravel 
(>2 mm) and rocks. Inorganic components exert the major effect on most soil prop- 
erties and on the overall suitability of soil as a plant growth medium. Organic com- 
ponents include plant and animal residues at various stages of decomposition, cells 
and tissues of soil organisms, and substances synthesized by the soil population. Or- 
ganic components, although usually present in much smaller amounts than inorganic 
components, affect soil properties significantly because of their high reactivity. 

Most inorganic soil components are crystalline compounds of definite structure 
called minerals. The sand and silt fractions are largely primary minerals, minerals 
formed at elevated temperatures and inherited unchanged from igneous and meta- 
morphic rocks, sometimes after passing through one or more sedimentary cycles. 
Primary minerals occur in the clay fraction of weakly weathered soils but are mi- 
nor constituents of the clay fraction of most agricultural soils. The most abundant 
primary minerals in soils are quartz (S1O2) and the feldspars (MAISiiOs), where M 
is a combination of Na' 1 ", K + , and Ca 2+ . Micas, pyroxenes, amphiboles, and other 
primary minerals are also common, but in smaller amounts. 

Minerals of the clay fraction of soils are largely secondary, that is, formed by low- 
temperature reactions and either inherited from sedimentary rocks or formed directly 
in the soil by weathering. These secondary {authigenic) minerals in soils commonly 

129 



1 30 INORGANIC SOLID PHASE 

include the layer silicates, Al and Fe hydroxyoxides, and carbonates and sulfur com- 
pounds. Early workers assumed that clay minerals formed in soils were amorphous, 
noncrystalline spheres. X-ray diffraction revealed that many clay minerals are in- 
stead crystalline. The aluminosilicate clay minerals are layered, similar to mica, and 
have the same relative dimensions as a stack of postage stamps. 

Other important constituents of the clay fraction are the so-called free oxides. 
These are Al, Fe, Mn, and Ti hydroxyoxides that accumulate in the soil as weath- 
ering removes silicon. The free oxides range from amorphous to crystalline and are 
often the weathered outer layer of soil particles. The hydroxyoxides, plus amorphous 
aluminosilicates such as allophane, are the most important clay-sized nonlayer min- 
erals in soils. 

The most abundant carbonate in soils is calcite (CaCCh), although pure calcite 
rarely precipitates directly from the soil solution. Calcium carbonate is common in 
semiarid and arid region soils and is often present in humid region subsoils derived 
from calcareous parent material. Ca carbonate accumulates in loose and porous to 
strongly indurated and rock-like layers in semiarid and arid soils. Gypsum (CaSC>4 • 
2H2O) occurs in semiarid and arid region soils. The major sulfur mineral is pyriie 
(FeS2). Pyrite is frequently associated with shales and coal beds and may form in 
soils under reducing conditions. 



5.1 CRYSTAL CHEMISTRY OF SILICATES 

Layered aluminosilicates are the most important secondary minerals in the clay frac- 
tion of soils. When layer silicate minerals are clay or colloidal size (<2 p,m effec- 
tive diameter), their large surface area greatly influences soil properties. Most of the 
important clay minerals have similar silicate structures. Inasmuch as clay minerals 
are such important clay components, and as different clay minerals can change soil 
properties greatly, an understanding of soil properties begins with an understanding 
of silicate structures. 

When atoms combine, the bond between them changes the electron distribution 
from that of the atomic state. The type of bond depends on the electronic structure 
of the combining atoms. Ionic or electrostatic bonding occurs between oppositely 
charged ions such as Na + and CI - . Such ions are formed by the complete loss or 
gain of electrons to form positive or negative ions having an electron structure like 
an inert gas. The formation of Na + and Cl~ from Na and CI atoms is, for exam- 
ple, 

:Cl- + e~ ->■ :C1:~ Na- -> Na~ + e" 

where the dots denote electrons in the outermost, or valence, shell. Alkali and alka- 
line earth metals and the halogens tend to gain or lose electrons readily and are the 
most likely to form ionic bonds. 

Ionic bonding is strong, and ionic-bonded compounds tend to be hard solids and 
have high melting points. Ionic bonding is also undirected — exerted uniformly in all 



directions. The valence of a given ion is shared by the surrounding ions of opposite 
charge. The number of such neighbors is determined by their size relative to the size 
of the central ion. Ionic bonds predominate in many inorganic crystals, including the 
silicates. 

Covalent bonding (shared electron pairs) is common between identical atoms or 
atoms having similar electrical properties, such as in H2O, F2, CH4, and C (dia- 
mond). In covalent bonding the electrons are shared between atoms so that each 
atom attains the inert gas electronic structure. For example, 

H 

H:'6: H:C:H 
H H 

Covalent bonding is strong, but directional. Bond angles in covalently bonded struc- 
tures are determined by the geometric positions of the electron orbitals (orbits) in- 
volved. Covalently bonded molecules have little tendency to ionize. Bonding within 
ionic radicals, or complex ions, such as SO;j~ , is frequently covalent. 

Hydrogen bonding occurs between H + and ions of high electronegativity, such 
as F~, O 2- , and N 3 ~. The hydrogen bond is essentially a weak electrostatic bond 
and is important in crystal structures of oxy compounds, such as the layer silicates. 
Summed over many atoms, the individually weak hydrogen bonds can strongly bond 
adjacent structures. 

The weak electrostatic force between residual charges on molecules is van der 
Waals bonding. Residual charges may result from natural dipoles of unsymmetrical 
molecules, polarization dipoles, or vibrational dipoles. These van der Waals forces 
are generally obscured by stronger ionic and covalent bonds but may dominate the 
properties of some molecules. 

Each type of chemical bond imparts characteristic properties to a substance. If 
more than one type of bond occurs in a crystal, the physical properties such as hard- 
ness, mechanical strength, and melting point are generally determined by the weakest 
bonds. These are the first to yield under mechanical or thermal stress. Thus, the phys- 
ical properties of layer silicates are determined largely by the strength of the bonds 
between their layers. 

Although differences in the types of bonds may seem clear-cut, bonding in most 
crystals is somewhere in between. For example, the Si-0 bond in the silicates is 
intermediate between purely ionic and purely covalent bonding. The degree of ionic 
nature of the Si-O bond is sufficient, however, to apply the rules for ionic bonding 
to silicate structures. 

Bonding within the silicate layers is predominantly ionic. As a result, forces are 
undirected and ion size plays an important role in determining crystal structure. 
Table 5. 1 shows the crystal radii of common ions in silicates. The distance between 
two adjacent ions in a crystal can be measured accurately by x-ray methods. From 
a series of such measurements between different ions, the effective contributing ra- 
dius of each ion can be determined. An ion has no rigid boundary; an ion's radius 
depends on the number of its orbital electrons and on their relative attraction to the 
ion's nucleus. The radius of Fe ions, for example, decreases from 0.074 to 0.064 nm 



Table 5.1 . Crystal ionic radii of selected cations, and 
coordination numbers for cations with oxygen 





Crystal 


Coordination Number 




Ionic Radius 
(nm) 


with 


Oxygen 




Ion 


Observed 


Predicted 


Si 4 + 


0.042 


4 




4 


Al 3 + 


0.051 


4, 6 




4 


Fe 3+ 


0.064 


6 




6 


Mg 2+ 


0.066 


6 




6 


Fe 2+ 


0.074 


6 




6 


Na + 


0.097 


6. 8 




6 


Ca 2 + 


0.09 


8 




6 


K+ 


0.133 


8, 12 




8 


NH+ 


0.143 


8, 12 




8 


o 2 - 


0.132 


— 




— 



"Reprinted with permission from Handbook of Chemistry ami Physics, 
50th ed. Chemical Rubber Co. Inc. Cleveland. OH (1969-1970). 



as the valence changes from Fe(II) to Fe(IIl). The ion radius also depends on the 
configuration of the ion structure. 

The ion radivis of oxygen (O 2- ) is much larger than that of most cations found 
in silicates. The oxygen ion constitutes 50-70% of the mass, and over 90% of the 
volume, of most common silicate minerals. Silicate structures are largely determined 
by the manner in which the oxygen ions pack together. 

An ion surrounds itself with ions of opposite charge. The number of anions that 
pack around a central cation depends on the ratio of the cation and anion radii and 
is called the coordination number of the central ion. Because O 2- is virtually the 
only anion in soil minerals, our interest centers on the different cations. Assuming 
that ions act as rigid spheres, the stable arrangements of cations and anions can be 
calculated from the packing geometry of their crystal radii (Table 5.2). 



Table 5.2. Spatial arrangement of rigid spheres in relation to radius ratio and 
coordination number 







Coordination 


Radius Ratio 


Arrangement of 


Number of 


' 'cation /''anion) 


Anions Around Cations 


Central Cation 


0.15-0.22 


Corners of an equilateral triangle 


3 


0.22-0.41 


Corners of a tetrahedron 


4 


0.41-0.73 


Corners of an octahedron 


6 


0.73-1 


Corners of a cube 


8 


1 


Closest packing 


12 



CRYSTAL CHEMISTRY OF SILICATES 1 33 

Ions are held together rigidly in a crystal structure, as determined by geometry 
and by electrical stability. More than one structure may meet the necessary require- 
ments, but the most stable form will be the one having the lowest potential energy. 
The requirement of electrical neutrality means that the sum of positive and negative 
charges must be zero. Ions of opposite charge do not pair off to achieve neutrality. 
Instead, the cation's positive charge is divided among surrounding anions. The num- 
ber of oxygen ions around each cation is determined by the coordination number, or 
radius ratio, of the cation and O 2- , rather than by the charge of the cation. 

Table 5.2 shows the predicted and observed coordination numbers of common 
cations with 2 ~. The Si 4+ cation occurs in fourfold or tetrahedral coordination. 
Aluminium is generally found in sixfold or octahedral coordination but also occurs 
in tetrahedral coordination in igneous minerals. Where the radius ratio is near the 
boundary between two types of coordination, the cation may occur in either coordi- 
nation, depending on conditions during crystallization. High crystallization temper- 
atures generally favor low coordination numbers. In high-temperature minerals Al 
tends to assume fourfold coordination and to substitute for Si. At lower temperatures 
Al tends to occur in sixfold coordination and Al substitution for Si is rare. 

The tetrahedral and octahedral units formed around Si and Al are basic to the 
structures of silicate minerals. Crystallography developed centuries ago, before ions 
were known. Crystallographers very cleverly deduced that crystals were formed by 
the packing of simple structures like tetrahedra and octahedra. Since they did not 
know about ions, their concern was about shapes. The number and arrangement of 
ions in a structure is more fundamental than the number of faces of the structure, but 
the old crystallographic nomenclature persists. The tetrahedral structure (Fig. 5.1a) 
is four O 2 " ligands coordinated around one Si 4+ , giving the unit Si0 4 ~. The elec- 
trostatic bond strength (ion charge divided by the number of bonds to the ion) for 
the tetrahedral unit is 1 . In fourfold coordination, the hole between the four O 2- ions 
is 0.225 times the O 2- radius, or 0.030 nm, if the ions are rigid spheres. In reality, 
cations ranging from 0.029 to 0.052 nm radius occur in fourfold oxide coordination. 
The radius of Si 4+ in fourfold coordination is about 0.042 nm, indicating that the 
ions are not completely rigid spheres. 

The octahedral (eight-sided) structure is formed by six anions coordinated around 
a central cation (Fig. 5.1b). The electrostatic bond strength is 1/2 if the central ion 
is trivalent, or 1/3 if the central ion is divalent. Ions commonly found in octahedral 



O 



CX-o^o 




©-silicon @ = olurmnum t mognesium, iron 

O = oxygens O =n yd r oxyls 

<o> (b) 

FIGURE 5.1. Diagram of (a) a silica tetrahedron, and (b) an octahedron of aluminum, magne- 
sium, or iron. 



1 34 INORGANIC SOLID PHASE 




FIGURE 5.2. Silica tetrahedron showing position and relative size of oxygen O and silicon 
ions. 



coordination in layer silicates are Al 3+ (radius 0.05 1 nm in this coordination), Mg 2+ 
(0.066 nm), and Fe 2+ (0.074 nm). Figure 5. 1 and all subsequent drawings of mineral 
structures greatly exaggerate bond lengths to point out structural features. Also, the 
size of the O 2- ions is reduced to show the relative positions of each ion. Figure 5.2 
shows a silicate tetrahedron more accurately. 

Minerals vary widely in chemical composition. Substitution of one element for 
another in mineral structures is common; pure minerals are rare in nature. Isomor- 
phic substitution, isomorphism, atomic substitution, and solid solution all refer to 
the substitution of one ion for another without changing the structure of the crystal. 
Such substitution takes place during crystallization and does not change afterward. 
Isomorphic substitution can occur between many ions of the same charge, but the 
size of the ions, rather than the charge, is more important than the charge. Electri- 
cal neutrality is maintained by simultaneous substitution of ions elsewhere in the 
structure, or by retaining ions on the outside of the structure. Isomorphic substitution 
generally takes place only between ions differing by less than about 10 to 15% in 
crystal ionic radii. 

The more common isomorphic substitutions in silicate structures are Al 3+ for 
Si 4+ in tetrahedral coordination, and Mg 2+ , Fe 2+ , and Fe 3+ for Al 3+ in octahedral 
coordination. Substitution between ions of unequal charge in layer lattice silicates 
leaves negative or positive charges within the crystal that are neutralized by ions on 
the surface of the lattices. The common substitutions in soils produce a net negative 
charge and contribute to the cation exchange capacity of soils. Some areas of soil 
clay particles have a net positive charge and an anion exchange capacity. The anion 
exchange capacity is usually less than the cation exchange capacity and may result 
from Ti 4+ substituting for Al 3+ . 

A crystal is an arrangement of ions or atoms that is repeated at regular intervals 
in three dimensions. The smallest repeating three-dimensional array of a crystal is 
called the unit cell. The unit cell dimensions a and b (the x and y dimensions) are 
constant for a given mineral; the c (or z) dimension is also constant, except for the 
special case of swelling layer silicates (Fig. 5.3). The basal plane is the a-b plane. 
The chemical composition of layer silicate minerals is normally expressed as one- 
half of a unit cell, to simplify the chemical formulas. 



STRUCTURAL CLASSIFICATION OF SILICATES 1 35 




FIGURE 5.3. The unit cell of a crystal. It is a parallelepiped with angles or, p, and y and edges 
a, b, and c. Positions of atoms in the crystal are usually given as x, y, and z coordinates scaled 
as fractions of the corresponding cell edges a, b, and c. (a) Cell for kaolinite. (b) The outline of a 
crystal of kaolinite showing the orientation of one of its unit cells. The unit cell is a = 0.515 nm, 
b = 0.85 nm, c = 0.715 nm, a = 91.8°, p = 104.8°, and y = 90.0°. The crystal is about 
9.0 nm wide, 10.3 nm long and 2.1 nm thick, about 10 cells by 20 cells by 3 cells, or 600 cells 
in all. This crystal is at the smaller end of the range of kaolinite crystals found in soils. (From 
D. S. Greenland and U. H-B. Hayes, eds. 1 978. Chemistry of Soil Constituents. Wiley, New York.) 



5.2 STRUCTURAL CLASSIFICATION OF SILICATES 

The Si-O bond is so strong that the tetrahedral arrangement of four oxygen anions 
about the silicon cation appears to be universal in silicate structures. Different silicate 
structures arise from the various ways in which the S1O4 tetrahedra combine with one 
another. Silicate structures range from single, separate tetrahedra to those in which 
all corners of the tetrahedron are linked through oxygen to other Si04 tetrahedra. 
Table 5.3 gives a structural classification of the silicates. This chapter is limited to 
the silicate structures distinctive to soils. 

5.2.1 Layer Silicates 

Layer silicates, sheet-like phyllosilicates such as the familiar micas, are in primary 
rocks and in soils. The soil minerals are often called clay minerals. Since other com- 
ponents can also be in the clay fraction, layer silicates is a more accurate term. A 
typical layer silicate is a combination of a layer of AI-, Mg- or Fe(II)-0 octahedra 
plus one or two layers of Si-0 tetrahedra. The tetrahedral and octahedral sheets bond 
together by sharing oxygens at the corners of the tetrahedra and octahedra. Layer sil- 
icate minerals are differentiated by (l) the number and sequence of tetrahedral and 
octahedral sheets, (2) the layer charge per unit cell. (3) the type of interlayer bond 



136 



INORGANIC SOLID PHASE 



Table 5.3. Structural classification of silicates 3 



Structural Type Classification 



Structural 
Arrangement 



Si:0 
Ralio 



Mineral 
Examples 



Single tetrahedra Nesosilicates 

Disilicates Sorosilicates 

Ring structures Cyclosilicates 

Single chains Inosilicates 

Double chains Inosilicates 



Sheet structures Phyllosilicates 
(layer silicates) 



Framework 
structures 



Tectosilicates 



Single tetrahedra 


1:4 


Olivine, garnet 


Two tetrahedra sharing one 


2:7 


Hemimorphite 


comer 






Closed rings of tetrahedra 


1:3 


Beryl 


sharing two oxygens 






Continuous single chains of 


1:3 


Pyroxene 


tetraheda sharing two corners 




(augite) 


Continuous double chains of 


4:11 


Amphiboles 


tetrahedra sharing alternately 




(hornblende) 


two and three oxygens 






Continuous sheets of tetra- 


2:5 


Micas, niont- 


hedra each sharing three 




morillonite 


oxygens 






Continuous framework of 


1:2 


Quartz, 


tetrahedra each sharing all 




feldspars, 


four oxygens 




zeolite 



"Reprinted with permission from James D. Dana. 1959. Manual of Mineralogy, 17th ed. <C. S. Hurlbut, 
ed.). Wiley, New York. 



and interlayer cations, (4) the cations in the octahedral sheet, and (5) the type of 
stacking along the c dimension. 

The structural unit of the kaolin group is formed by superimposing a tetrahedral 
sheet on an octahedral sheet. Such minerals are referred to as /: / layer silicates. The 
top oxygens of the tetrahedral sheet are shared by the octahedral sheet, forming a 




H- bonding 
between basol planes 



c-spoang 
0.72 nm 



q- <y m 



s ; x t 
* si**.! 

V 9 * 



OH 
AJ 
O.OH 

Si 


Koolinite Al 2 Si 2 5 (0H) 4 

FIGURE 5.4. Schematic structure of kaolinite. (From F. E. Bear, ed. 1964. Chemistry of the Soil. 
ACS Monograph Series No. 1 60.) 



STRUCTURAL CLASSIFICATION OF SILICATES 1 37 

common plane of oxygen ions within the structure (Fig. 5.4). In the shared plane, 
two-thirds of the oxygen ions are shared between Si and Al. The other one-third of 
the oxygen ions have their remaining charge satisfied by H + . The upper surface 
of kaolin is a layer of closery packed OH groups. The bottom surface is composed of 
hexagonally open-packed oxide ions, with an OH recessed within the hexagonal 
(ditrigonal) oxygens. The I : l minerals are apparently very inflexible in their struc- 
tural requirements and allow little or no isomorphous substitution. The structure is 
electrically neutral. The l : l layers are held together strongly by the many hydrogen 
bonds between the OH groups of one sheet and the O ions of the next sheet. 

The 2:1 layer silicates are made up of an octahedral sheet sandwiched between 
two tetrahedral sheets. These unit layers then stack in the c direction. Figure 5.5 
shows the atomic arrangement of pyrophyllite, an electrically neutral 2:1 mineral. 
Soil 2: 1 layer silicates, on the other hand, have the same structure, but have extensive 
isomorphic substitution, which leads to largely negative charges within the crystal. 
This charge must be balanced by other cations, either inside the crystal or outside 
the structure. The magnitude of charge per formula unit, when balanced by cations 
external to the unit layer, is called the layer charge. The 2:1 minerals are classified 
according to layer charge in Table 5.4. 

The magnitude of layer charge plays a dominant role in determining the strength 
and type of bonding between the 2: 1 layers. If the layer charge is zero, as in pyro- 
phyllite, the 2:1 layers bond together by very weak van der Waals forces. If the layer 
charge is negative, the 2:1 sheets bond electrostatically and more strongly because 
cations enter between the unit layers. The greater the layer charge, the more cations, 
and the stronger the interlayer bond. Smectites (which include montmorillonite and 
bentonite) of low layer charge bond weakly so polar molecules, such as water, can 
enter between the sheets, and the minerals expand or swell as the soil becomes wet. 
In minerals of high layer charge, such as the micas {muscovite and biotite), the ionic 
bond of K + between the sheets is so strong that polar molecules cannot enter, the 
minerals are nonswelling (nonexpanding), and the soil does not shrink and swell with 
changing moisture content. Vermiculites are intermediate in layer charge and also in- 



von der Wools bonding 
between bosol plones 

c-spacmg [ T T 



0.93 nm 



C3-^ 9-^ „-• ^-A^ o-s.^-* °' 0H 

O" " 6""'"^> """V" t^""^* °' 0H 

Pyrophyllite fll 2 S14 O i0 (0H) 2 

FIGURE 5.5. Schematic structure of pyrophyllite. (From F. E. Bear, ed. 1 964. Chemistry of the 
Soil. ACS Monograph Series No. 160.) 



138 



INORGANIC SOLID PHASE 



Table 5.4. Typical 2:1 layer silicate minerals 





Layer Charge per 
Formula Unit in 


Predominant 
Octahedral Cation 


Mineral 
Group 


Tetrahedral 
Sheet 


Octahedral 
Sheet 


Al 3 + 
(Dioctahedral) 


Mg 2+ 
(Trioctahedral) 


Pyrophyllite-talc 
Smectites 

Vermiculites 
Micas 




0.25-0.6 



0.6-0.9 

1 




0.25-0.6 




Pyrophyllite 

Beidellite 

Monlmorillonite 

Vermiculite 

Muscovite 


Talc 

Saponite 

Hectorite 

Vermiculite 

Biotite" 



"Mg 2+ and Fe 2+ in octahedral coordination, K + in the intcrlayer position. 



termediate between micas and the smectites in their swelling properties. Within a 
given mineral group, specific minerals are defined by the predominant ion in the oc- 
tahedral coordination. Table 5.4 gives names for common minerals having Al 3 " 1 " in 
octahedral coordination. 

The 2: 1 layer silicate minerals are sometimes defined on the basis of the num- 
ber of octahedral positions occupied by cations. When two-thirds of the octahe- 
dral positions are occupied, such as in pyrophyllite (Al2Si40io(OH)2), the min- 
eral is call dioctahedral; when all three positions are occupied, such as in talc 
(Mg 3 Si40io(OH)2), the mineral is called trioctahedral. This difference in compo- 
sition of layer silicates can be fairly easily determined by x-ray diffraction, because 
each substitution slightly changes the dimensions of the unit cell. 

Chlorites are closely related to the micas and have about the same layer charge. 
In chlorite, the interlayer K + of mica is replaced by a positively charged octahedral 
brucite (Mg 3 (OH)6) sheet. The brucite sheet develops a positive charge when the 
Mg 2 " 1 " is partially replaced by Al 3 " 1 ". Such replacement is often about one-third of 
the Mg 2+ positions, to give the basic unit (Mg 2 Al(OH)6) + that fits into the inter- 
layer position of 2: 1 layer silicates to yield the 2: 1 : 1 chlorite (Fig. 5.6). Chlorites are 
nonexpanding, have low cation exchange capacities, and are generally trioctahedral. 



5.2.2 Relation of Structure to Physical and Chemical Properties 

The interlayer bond has a big effect on the physical and chemical properties of layer 
silicates. Bonding within the unit layers is much stronger than between adjacent unit 
layers. When the mineral is subjected to physical or thermal stress, it fractures first 
between the unit layers, along the basal plane. This is the reason for the flake-like 
shape of most macroscopic layer silicate crystals. Also, the stronger the interlayer 
bond, the greater the crystal growth in the c dimension before fracture. Hence, the 
size and shape of layer silicate crystals is a direct consequence of the strength of their 
interlayer bonds. 

The surface area of layer silicates is related to their expanding properties, and 
may be either external only or external plus internal. External surface refers to the 



STRUCTURAL CLASSIFICATION OF SILICATES 



139 



c -spacing 

0.95 to 1 .8 nm 

or more 



^^^^W%r o 



Montmorillonite No, (Ai 2 .,Mg x ) Si40io(OH)2 



Na 


Si 



0, OH 
Al, Mg 
0.0H 

Si 




FIGURE 5.6. Schematic structure of chlorite. (From F. E. Bear, ed. 1964. Chemistry of the Soil. 
ACS Monograph Series No. 160.) 



faces and edges of the whole crystal; internal surface is the area of the basal plane 
surfaces of each unit layer. Nonexpanding minerals exhibit only external surface; 
expanding minerals have both internal and external surface. The total surface area of 
montmorillonite can be as large as 800 x 10 3 m 2 kg -1 . The surface area of kaolinite, 
a nonexpanding and 1:1 layer silicate, is usually only 10 to 20 x 10 3 m 2 kg - ' . 

The c spacing of the layer silicates is determined by (1) the number of O-OH 
sheets per unit structure in the c dimension, and (2) the presence of ions and/or 
polar molecules in the basal plane. This spacing is conveniently measured by x-ray 
diffraction. Pyrophyllite has an electrically neutral lattice and hence no interlayer 
cations. Adjacent units of this mineral approach one another closely and form van 
der Waals bonds, preventing water entry. The resultant constant c spacing is 0.93 nm 
(Fig. 5.5). In minerals having weak interlayer bonds such as montmorillonite, cations 
and water or other polar molecules can enter between the basal planes, causing the 
c spacing to increase. The expansion varies greatly with the amount and type of 
polar molecule. In minerals with strong interlayer bonding, such as mica, chlorite and 
kaolinite, water, and other polar molecules cannot enter between the basal planes. 

Adsorbed cations are held by layer silicates to balance the negative charge re- 
sulting from isomorphic substitution and from unsatisfied bonds on crystal edges. 
The magnitude of the exchange capacity of the crystal edge is related to the number 
of unsatisfied bonds, and therefore is a direct function of crystal size. In 2:1 layer 
silicates, the larger portion of the cations balancing this mostly negative charge is 
in the basal plane. Where the layer charge is high, such as in mica, the bond en- 
ergy is so great, the adjacent layers are so tightly "collapsed," and the K + fits so 
well into the hexagonal holes in the basal plane that the adsorbed cations are not ex- 
changeable. The adsorbed cations of smectite minerals, on the other hand, are readily 
exchangeable. For vermiculite, Ca 2+ , Mg 2+ , and Na + ions in interlayer position are 
exchangeable, but K + and NH^" ions, because of their good fit, are not exchangeable 
by ordinary procedures. Thus, exchange capacity is directly related to layer charge 
until the charge becomes so large that the adsorbed cations cannot be removed. 



1 40 INORGANIC SOLID PHASE 

Charged areas on the mineral surface that arise from unsatisfied bonds and iso- 
morphic substitution help to retain polar molecules. In addition, the asymmetrical 
distribution of orbital electrons in O and OH groups produces local negative and 
positive (polar) areas. Both the charged and the polar surfaces actively adsorb polar 
molecules by hydrogen bonding and by van der Waals forces. 



5.3 SOIL LAYER SILICATES 

Soil clay minerals often differ appreciably from those of the pure minerals. Soil 
clays are usually less well ordered and smaller in size than pure minerals. Soil clay 
minerals of different composition and structure often overlap. Neighboring particles 
or sheets and interleafings and interstratifications of different layer silicates are com- 
mon. The mineralogy of soil clays is rarely simple or uniform because of appreciable 
isomorphic substitution. Coatings of Fe and Al oxides and organic matter on most 
layer silicates in soils further complicate the mineralogy, identification, and prop- 
erties of soil clays. Such coatings can decrease the cation exchange capacity, sur- 
face area, swelling, and collapse of expansible minerals. Oxide coatings, however, 
increase anion exchange and other properties associated with positively charged sur- 
faces. The minerals below dominate the clay fraction of most soils. 

5.3.1 Kaolins 

Kaolinite (Al2Si20s(OH)4) is the 1:1 layer silicate mineral that typifies the kaolins 
(Fig. 5.4). Kaolinite is common in soils, as hexagonal crystals with an effective di- 
ameter of 0.2 to 2 ixm. Si 4+ is apparently the only cation in the tetrahedral sheet 
of kaolinite, but Al 3+ or Mg 2H ' may occupy the octahedral positions. When Al 3+ 
is in the octahedral sites, the mineral is kaolinite or one of its poorly crystallized 
polymorphs, dickite or nacrite. When Mg 2+ is in octahedral sites, the kaolin mineral 
is antigorite (Mg 3 Si205(OH)4). Halloysite is a form of kaolinite in which water is 
held between structural units in the basal plane, yielding a c spacing of 1 .0 nm when 
fully hydrated. Most kaolin units, however, are held together in the basal plane by 
hydrogen bonding between oxygen ions of the tetrahedral sheet and OH ions of the 
octahedral sheet. 

Such hydrogen bonding prevents expansion (swelling, entry of polar molecules 
between unit layers) of the mineral beyond its c spacing of 0.72 nm. Surface area is 
limited to external surfaces and hence is relatively small, ranging from 10 to 20 x 
10 3 m 2 kg -1 . Kaolinite is a coarse clay with low colloidal activity, including low 
plasticity and cohesion, and low swelling and shrinkage. 

The unit formula for kaolinite has a Si/Al ratio of 1. This ratio is matched by 
its chemical composition, which suggests that soil kaolinites have little or no iso- 
morphic substitution. Any differences from 1 could be due to surface coatings that 
were not removed during preparation of the sample. Most of the 1 0- to 1 OO-mmol(-f-) 
kg -1 cation exchange capacity of kaolinite has been attributed to dissociation of OH 
groups at clay edges. However, if only one Si 4+ of every 200 in the silica sheet were 



SOIL LAYER SILICATES 



141 



substituted by Al 3+ , the net negative charge would be 200 mmol(-) kg -1 . Chemi- 
cal analysis of clays is not sufficiently sensitive to prove or disprove this extent of 
isomorphic substitution. The cation exchange capacity of kaolinite is highly pH de- 
pendent, suggesting that OH dissociation is the predominant source of charge rather 
than isomorphic substitution. 

5.3.2 Smectites (Montmorillonite) 

Smectites are 2: 1 layer silicates with layer charge of 0.25 to 0.6 per formula unit. 
An idealized formula for smectite is KAI7S11 |C>3o(OH)6, but much isomorphic sub- 
stitution occurs. Because of the relatively low layer charge compared to mica and 
vermiculite, smectites expand freely. The c spacing varies with the exchangeable 
cation and the degree of interlayer solvation. Complete drying yields a spacing of 
0.95 to 1 .0 nm, and full hydration can swell the layers up to lens of nanometers. 

Depending on the predominant octahedral cation and on the location of isomor- 
phic substitution, several names have been assigned to minerals within the smectite 
group (Table 5.4). The predominant one is montmorillonite in which Al 3+ is the 
major, and Mg 2 "'" the minor, octahedral cation (Fig. 5.7). A typical half-unit-cell for- 
mula is Na A ((Al2-.. v Mg v )Si40|o(OH)2) in which Na + is the charge-compensating 
exchangeable cation, Al and Mg are in the octahedral layer, and Si is in the tetra- 
hedral layers. Soil montmorillonites often exhibit imperfect isomorphic substitution, 
with some Al 3+ substituting for Si 4+ in the tetrahedral sheet and Fe 2+ and Mg 2+ 
substituting for Al 3+ in the octahedral sheet. 

Cation exchange capacities for montmorillonite range from 800 to 1200 mmol(+) 
kg -1 . The cation exchange capacity is only slightly pH dependent. The low layer 
charge allows the mineral to expand freely, exposing both internal and external sur- 
faces. Such expansion yields a total surface area of from 600 to 800 x I0 3 m 2 kg -1 , 



y^—o-^\^^ 



a q,o 
or 6 X) 



H 2 

Mg 

H 2 



c-spocing 
1.0 to 1.5 nm 



y^y^^V^ «•* 



A A. J9^ -<X A -# o.oh 

-^ 0,OH 
Al.Si 
Vermiculite IMglH ? 0) 6 l» KMg. Fe" 'l 3 (Si 4 „„. Al„| O, tOH) 2 ] 

FIGURE 5.7. Schematic structure of montmorillonite. (From F. E. Bear, ed. 1964. Chemistry of 
the Soil- ACS Monograph Series No. 160.) 




142 



INORGANIC SOLID PHASE 



with as much as 80% of the total due to internal surfaces. Montmorillonite has 
high colloidal activity, including high plasticity and cohesion, and high swelling and 
shrinkage. Montmorillonite normally occurs as a fine clay with irregular crystals hav- 
ing an effective diameter of 0.01 to 1 /xm. Smectites are common in Vertisols and in 
soils of alluvial plains. 

5.3.3 Vermiculites 

Vermiculites occur extensively in soils formed by weathering or hydrothermal alter- 
ation of micas. The layer structure of vermiculite resembles that of the mica from 
which the mineral is derived (Fig. 5.8). Both trjoctahedra) and dioctahedral vermi- 
culites exist. Weathering or alteration of the precursor micas replaces the interlayer 
K + mostly with Mg 2+ and expands the c spacing to 1.4-1.5 nm. 

The name vermiculite includes several minerals. For our purposes, vermiculite 
refers to a 2:1 layer silicate capable of only limited expansion, having Al 3+ 
substituted for Si 4+ in the tetrahedral sheet to the extent of 0.6 to 0.9 per unit 
formula, and with Mg 2+ and Fe 2+ as the octahedral cations. An idealized half- 
unit-cell formula is (Mg(H 2 0) G )„/2((Mg, Fe) 3 (Si 4 _„, Al, i ))Oi (OH) 2 , where the 
hydrated Mg(H20) 6 + is the exchangeable cation. An interesting property of some 
vermiculites is the internal balancing of layer charge originating from the parent 
mica by substituting Fe 3+ or Al 3 ' 1 ' for Mg 2+ and Fe 2+ in the octahedral sheet to 
yield Nao.7(Alo.2Fe(ni) .4Mg 24 )(Si2.7Ali.3)Oi (OH)2. Isomorphic substitution in 
the tetrahedral sheet yields a charge of —1.3 per unit cell. Isomorphic substitution in 
the octahedral sheet yields a charge of +0.6 in the unit cell. The net charge of —0.7 
per unit formula is balanced in this example by Na + . 

The layer charge in vermiculite gives rise to a cation exchange capacity of from 
1200 to 1500 mmol(+) kg -1 , which is considerably higher than the exchange ca- 
pacity of montmorillonite. As with montmorillonite, the cation exchange capacity is 
only slightly pH dependent. Vermiculite swells less than montmorillonite because of 
its higher layer charge. The mineral is nonswelling (with a c spacing of 1 .0 nm) when 
saturated with K + or NH^ ions. Such ions are termed fixed because they cannot be 



T 



c-spocmg 
1.0 nm 







K 

Si, Al 



JD o.oh 

I 

■zi A8 

y a- X> y T ^° 0,0H 

M.co K[A!2(SijAl)Oio(OH) 2 ] 



Si, AB 
O 



FIGURE 5.8. Schematic structure of vermiculite. (From F. E. Bear, ed. 1964. Chemistry of the 
Soil. ACS Monograph Series No. 160.) 



CHLORITES 



143 



T 



c-spocing 
1.0 nm 





K 

Si, Al 



^£> 0.0H 

;4c y< x > as 



Mica K[AS 2 (Si 3 Al)Oio(OH) 2 ] 



Si, AJ 




FIGURE 5.9. Schematic structure of mica. (From F. E. Bear, ed. 1964. Chemistry of the Soil. 
ACS Monograph Series No. 160.) 



exchanged with ordinary salt solutions. The total surface area of vermiculite, when 
not K + or NHJ saturated, ranges from 600 to 800 x 10^ m x kg - 1 . 



5.3.4 Micas 

Micas are abundant in soils as primary minerals inherited from parent materials. 
Micas are not known to form to any significant extent in soils. Micas are precursors 
for other 2:1 layer silicates, notably vermiculites. Micas are commonly present in 
soils as components of particles that have been partially transformed to expansible 
2:1 minerals. As a result, mica is often interstratified with other minerals. Altered 
mica containing less K + and more water than well-ordered mica is called hydrous 
mica (formerly Mite). 

A typical 'half-unit-cell formula for mica is KiAlilSij Al)Oio(OH')2>, and micas 
can be either dioctahedral or tri octahedral. Figure 5.9 shows the layer structure of the 
dioctahedral form, muscovite, which is clear and colorless. Isomorphic substitution 
of Al for Si in the tetrahedral layer creates negative charge close to the layer surface, 
which results in a strong coulombic attraction for charge-compensating cations- In- 
terlayer K + is so strongly adsorbed that it is not exchanged in standard cation ex- 
change capacity (CEC) determinations. Thus, despite the large layer charge (about 
- 1 per unit cell for many micas), the CEC is only 200 to 400 mmol(+) kg - ' . Total 
surface area is about 70 to 120 x 10 3 m 2 kg -1 and is restricted to external surfaces. 
Micas are nons welling and are only moderately plastic. The fixed K+ is released 
slowly during weathering and is a source of K + for plants. Biotite is the dark-colored 
mica, trioctahedral, with Fe and Mg in the octahedral sheet. 



5.4 CHLORITES 

Chlorites occur extensively in soils and are 2:1:1 layer silicates (Fig. 5.6). The posi- 
tively charged and substituted brucite sheet between the negatively charged mica-like 
sheets restricts swelling, decreases the effective surface area, and reduces the effec- 



1 44 INORGANIC SOLID PHASE 

tive cation exchange capacity of the mineral. An idealized half-unil-cell formula is 
(AlMg 2 (OH) 6 ). v (Mg3(Si 4 ^Al. v )Oi()(OH) 2 ). Substitution in such classical chlorites 
is in the tetrahedral layer of the 2:1 portion, with the brucite sheet serving as the in- 
terlayer "cation." The layer charge of the 2: 1 portion of the mineral varies but is 
similar to mica. Cation exchange capacities range from 100 to 400 mmol(+) kg -1 ; 
total surface areas range from 70 to 150 x 10 3 m 2 kg -1 . 

Chlorite is common in sedimentary rocks and in productive soils derived there- 
from. The elemental composition of chlorites varies widely, however, with chromium 
and nickel occurring in mafic (Fe- or Mg-containing) chlorites. Serpentine-derived 
soils contain chlorite and many are infertile because of their high Mg. and low Ca 
contents. 



5.5 ACCESSORY MINERALS 

As soils weather and Si, Ca, Mg, Na, and K are leached away, the soil's colloidal 
fraction becomes enriched with Al, Fe, Mn, and Ti oxides and hydroxyoxides. The 
structural organization of these hydroxyoxides ranges from amorphous to crystalline. 
These Al, Fe, and Ti oxides and allophane are prominent nonlayer silicate minerals 
in most soils and their content in soils increases with increased weathering. 

Most of the crystallized secondary minerals found in soils have passed through 
intermediate amorphous steps during chemical "weathering. Amorphous is defined as 
being nondetectable by x-ray diffraction. The distinction between amorphous and 
crystalline materials is vague, and various degrees of crystallinity can occur during 
the reorganization of hydrous gels that precipitate from the soil solution. Very small 
crystals can appear amorphous to many tests of crystallinity. Such materials have 
been termed cryptociystalline. 

5.5.1 Allophane and Imogolite 

Allophane is a general name for amorphous aluminosilicate gels. These are one of the 
more common groups of amorphous materials in soils. The composition of allophane 
varies widely but includes mostly hydrated AI2O3, Fe2C>3, and SiCb- Only minor 
amounts of Mg 2+ , Ca 2+ , K + , and Na + are generally present. The Al/Si ratio is 
usually between 1 and 2. Whether allophane is a mixture of individual Si or Al 
oxide gels or whether it is an amorphous hydrous aluminosilicate in which oxygen 
anions are shared between Si and Al ions is unclear. In any case, Si is apparently 
in tetrahedral coordination and Al is in octahedral coordination. Allophane was first 
noted in Japanese soils derived from volcanic parent materials that weather quickly, 
but it is probably present in most soils as an intermediate product of weathering. 

Allophane can have a high cation exchange capacity at neutral to mildly alkaline 
pH, perhaps 1 500 mmol(+) kg" ' , but the CEC is highly dependent on pH and degree 
of hydration. Values reported in the literature range from 100 to 1500 mmol(-f) kg -1 , 
depending on pH. The CEC measurement is indefinite because exchangeable cations 
are loosely adsorbed and hydrolyze extensively during washing with aqueous alcohol 



ACCESSORY MINERALS 145 

mixtures. Allophane has a high surface area, 70 to 300 x 10 3 m 2 kg" ' , but this also 
varies widely with degree of crystallinity and pH. 

Imogolite (Al2Si03(OH)4) was also first described in Japanese soils derived from 
volcanic ash. Although not highly crystalline, imogolite can be recognized by x-ray 
diffraction. Imogolite's and gibbsite's solubility are thought to determine the Al 3+ 
concentrations in the soil solution of many moderately acid soils. 

5.5.2 Zeolites 

Zeolites are three-dimensional aluminosilicate structures, like feldspars, in which 
tetrahedra are linked by sharing their vertices (Table 5.3). The tetrahedra are linked 
into 4-, 6-, 8-, and 12-membered rings, joined together less compactly than in the 
feldspars. Analcime (NaAlSiaOe • H2O) is a representative zeolite. The open frame- 
work of zeolites leaves channels of different sizes that run in several directions 
through the crystal. The channels contain loosely held water molecules and charge- 
balancing cations that are freely exchangeable. The channels frequently interconnect 
and are usually larger than the diameters of the common cations, so both water and 
charge-balancing cations diffuse readily through the crystals. Some zeolites have 
smaller-sized channels that effectively prevent movement of large molecules, lead- 
ing to the use of zeolites as "molecular sieves." 

5.5.3 Al, Fe, Ti, and Mn Hydroxyoxides 

Gibbsite (Ab(OH)6) is the most abundant Al hydroxide in soils, occurs in large 
amounts in highly weathered soils, 20-30% and more by mass, and is the stable low- 
temperature form of Al hydroxide. Crystalline gibbsite consists of sheets held to- 
gether by hydrogen bonding between adjacent hydroxy 1 ions arranged directly above 
and below one another. In acid soils, gibbsite and the Fe hydroxides react strongly 
with phosphate and are responsible for keeping phosphate unavailable to plants and 
for adsorbing SO^ - and reducing its availability to plants. Gibbsite and the Fe hy- 
droxyoxides are responsible for much of the pH dependence of soil CEC. These 
hydroxyoxides have a net positive charge in acid soils due to H + adsorption. As the 
pH increases, the charge changes from positive to negative as the H + dissociates and 
the amount of phosphate and sulfate retained decreases. Boehmite (AlOOH) occurs 
in intensively leached, highly weathered soils. It also can be formed from gibbsite by 
heating to about 130° C. Corundum (AI2O3) is a high-temperature form rarely found 
in soils. 

Hematite (Fe203) and goethite (FeOOH) are the most common Fe oxides found in 
soils. Hematite occurs in highly weathered soils and is pink to bright red. Goethite is 
also characteristic of strongly weathered soils and is brown to dark yellowish-brown. 
Both hematite and goethite occur as amorphous coatings on soil particles, imparting 
the red and brown colors characteristic of soils. The amorphous coatings transform 
to crystalline forms with aging, or as the amounts increase. As with gibbsite, the 
soil content can be 20-30% in highly weathered soils. Aluminium and iron ore de- 
posits are areas of extreme soil weathering. The crystallization of amorphous Fe 



146 INORGANIC SOLID PHASE 

hydroxyoxides is responsible for the irreversible hardening, upon drying, of plinthite 
(laterite) of tropical soils into stonelike materials. Chemically, the Fe hydroxyoxides 
behave similarly to gibbsite as described above. Magnetite (FejOa) is a magnetic Fe 
oxide inherited from the parent rock. It usually occurs as sand-sized grains of high 
specific gravity. Magnetite oxidizes to maghemite (Fe2C>3), which is also magnetic. 

The Ti oxides commonly found in soils and clay sediments are rutile and anatase, 
both T1O2 and inherited from the parent rock. Because Ti oxides resist weathering so 
strongly, they are often used as indicators of the original amount of parent material 
from which a soil has formed. 

Manganese oxides are a poorly understood and amorphous mixture of Mn(IlI) 
and Mn(IV). Pure pyrolusite (Mnd) is rare, a more accurate formula would be ap- 
proximately MnOi.g. Many transition metal ions have the same size as the Mn ions 
so isomorphous substitution is common. Hence, Mn hydroxyoxides can retain these 
other cations. For a time some workers believed that Mn hydroxyoxides were an 
important part of the soil's retention of trace metals. Per unit weight, this may be 
so. Soils, however, contain about 0.1% Mn hydroxyoxides, little compared to the 
amounts of Al and Fe hydroxyoxides. The manganese nodules that receive attention 
as a Mn ore are Mn-rich iron oxide nodules that are found on some ocean floors. 
Manganese nodules have not been found in soils. 



5.6 CHARGE DEVELOPMENT IN SOILS 

The two properties that most account for the reactivity of soils are surface area and 
surface charge. Surface area is a direct result of particle size and shape. Most of 
the total surface area of a mineral soil is due to clay-sized particles and soil organic 
matter. Charge development in soils is due to these same two fractions, although 
the sand- and silt-size fractions may contribute some cation exchange capacity if 
coarse-grained vermiculite is present. Charge development in soils occurs as a result 
of isomorphic substitution and of ionization of functional groups on the surface of 
solids, again primarily in the colloidal fraction, resulting in the permanent and the 
pH-dependent charges of soils. 



5.6.1 Permanent Charge 

Isomorphic substitution is the substitution of one ion for another of similar size 
within a crystal lattice. The substituting ion may have a greater, equal, or lower 
charge than the ion for which it substitutes. In layer silicate structures, cations can 
substitute for coordinating cations in either the tetrahedral or the octahedral sheets. 
If a cation of lower valence substitutes for one of higher valence, such as Mg 2+ for 
Al 3+ or Al 3+ for Si 4+ , the negative charge of O 2- and OH - ions in the mineral 
structure is left unsatisfied, yielding a net negative charge. Isomorphic substitution 
can also result in positive charge, by Al 3+ substituting for Mg 2+ in the brucite inter- 
layer of chlorite, but negative charge tends to predominate in soil minerals. 



CHARGE DEVELOPMENT IN SOILS 



147 



Isomoiphic substitution occurs during crystallization of layer silicate minerals in 
magmas and in soils. If the primary cation is unavailable as the unit cell forms, an- 
other cation can sometimes squeeze in. The resulting permanent charge is essentially 
independent of the soil solution composition surrounding the particle. Isomorphic 
substitution is the principal source of negative charge for the 2: 1 and 2:1:1 layer 
silicates, but is of minor importance for the 1:1 minerals. 

5.6.2 pH-Dependent Charge 

The total charge of soil particles varies with the pH at which the charge is measured. 
Figure 5.10 illustrates/?//- dependent charge, where some portion of the soil changes 
from positive charge at low pH to negative charge at higher pH. The soil's total charge 
is the algebraic sum of its negative and positive charges. The relative contribution of 
permanent and pH-dependent charge depends on the composition of soil colloids. 
Relatively young and weakly weathered soils characteristic of Europe and North 
America have a net negative charge, because of the higher pH and layer silicate and 
organic matter content of these soils. Highly weathered and volcanic soils, on the 
other hand, are dominated by allophane and hydrous oxides, may have a low pH, and 
may have a net neutral to positive charge (Table 5.5). Subsoils are usually lower in 
organic matter so the relative amount of negative charge relative to positive charge 
decreases. The zero point of charge (ZPC) is an index of the positive and negative 
charge on soil colloids. The ZPC is the pH at which negative and positive charges of 
a colloid are equal. The ZPC for the soil of Fig. 5.10 would be pH < 3. 

Crystal bonding ends at the particle-soil solution interface. At the particle edge, 
the charge of the structural cations and O 2- ions is not compensated by surrounding 
structural ions. Electrical neutrality is necessary and is maintained by interacting 
with H + , OH~, and water, and by adsorbing cations or anions from the soil solution. 
The primary source of pH-dependent charge is considered to be the loss of adsorbed 
H + and OH~ on inorganic solids and H + from organic acids, phenols, and other 
functional groups in soil organic matter. 



pH- dependent 




FIGURE 5.10. Representative change of positive and negative charges on soils with pH. (From 
W. D. Guenzi, ed. 1 974. Pesticides in Soil and Water. American Society of Agronomy, Madison, 
Wl.) 



1 48 INORGANIC SOLID PHASE 

The soil solids that contain functional groups capable of developing positive pH- 
dependent charge include layer silicates, allophane, hydroxyoxides, and organic mat- 
ter. In organic matter, the functional groups that create pH-dependent positive charge 
include hydroxyl ( — OH), carboxyl ( — COOH), phenolic ( — C6H4OH), and amine 
( — NH2). Equation 5.1 shows how an inorganic hydroxyoxide, an Al hydroxyoxide 
in this case, can change from negative to positive charge by adsorbing H + : 

(Al)-OH l/2+ + H + = (Al)-OH2 /2+ (5.1) 

The other hydroxyoxides are similar. The extent of the reaction is pH-dependent. 
This effect is accentuated by the tendency of these hydroxyoxides to be thin coat- 
ings on soil particles, which increases their activity per unit mass. Soils containing 
large amounts of Al and Fe oxides have a strongly pH-dependent charge and highly 
variable CEC. 

Figure 5.11 shows how pH-dependent charge develops at the crystal edges of 
kaolinite. Depending on the pH of the soil solution, the charge can be either positive 
or negative. Jackson suggested that the dissociation of H + occurred at pK A = 5.0 for 
the AI-OH2 group, 7.0 for (Al,Si)-OH, and 9.5 for Si-OH. The high pK s for Si-OH 
groups indicates that the deprotonation (H + loss) occurs only at high pH. The pH- 
dependent charge of layer silicates is more likely due to reversible protonation and 
deprotonation of Al-OH rather than Si-OH groups. 

The contribution of edge OH groups to pH-dependent charge is related to the 
acidity of the edge groups and to the area of edge surface. For 2: 1 minerals such as 
montmorillonite, the functional groups are apparently weakly acidic and dissociate 
only at high pH. In addition, the amount of edge surface for 2: 1 minerals is small 
relative to the basal (planar) surface. Kaolinite, on the other hand, tends to stack 
without swelling in the c dimension, increasing the edge area compared to the basal 
plane area. For both reasons, pH-dependent charge is more important for kaolinite 
than for smectites or vermiculites. As a rule of thumb, only 5 to 10% of the negative 
charge on 2: 1 layer silicates is pH dependent, whereas 50% or more of the charge 
developed on 1 : 1 minerals can be pH dependent. 

Figure 5.12 shows the pH-dependent charge in kaolinite. The CI - anion is re- 
tained by kaolinite in acid solutions, indicating the presence of positive sites, prob- 



OH 


OH 






Ot-D 


/ 


/ 






/ 


Si 


Si 






Si 


\ 


\ 






\ 


0H{*U2) ^± H»* 


OH (<-i<2) 


♦ 20H" 


■^- fc 


0(-"2) 


/ 


/ 






/ 


A« H 


A8 






AJt 


\ <»'/2) 


\ 






\ 


OH 


OH (-K2) 






OH(-if2) 


Acid 


Neutral pH 






Bosic 



'2H 2 



FIGURE 5.11. Representation of pH-dependent charge at kaolinite edges. (By permission from 
R. K. Schofield and H. R. Samson. 1953. Clay Miner Bull. 2:45.) 



CHARGE DEVELOPMENT IN SOILS 



149 






positive odsorption of Cr 
(electrosiotic Attraction) 



3. 



[ \ negative adsorption of Cr" 

i ° °" — -O — (electrosiotic repulsion) 



5 6 7 8 9 

pH 



FIGURE 5.12. Chloride adsorption by kaolinite at various pH values. (By permission from 
R. K. Schofield and H. R. Samson. 1953. Clay Miner Bull. 2:45.) 



ably Al-OH*. In basic solutions, the functional group changes to the negatively 
charged Al-OH~, which repels anions. The pH at which positive and negative 
charges are balanced, ZPC, for this kaolinite is indicated by the vertical dashed line 
at about pH 6.5. 

In highly weathered soils, Fe and Al oxides are abundant and can develop con- 
siderable pH dependent charge, as can Ti, Cr, and Mn oxides. For example, Fe 3 " 1 " in 
hematite is in sixfold coordination with O 2- . Each valence bond of an oxygen sup- 
plies —0.5 charge to the Fe ion. The remaining — 1 .5 charge of each O is satisfied 
by Fe cations in adjacent octahedra. At the soil solution interface (the edge of the 
crystal), however, the ion charges in the crystal have to be satisfied by I-I+ and OH~ 
ions in the soil solution. In effect, the ions in the crystal complete their coordination 
spheres by interacting with the soil solution. The result is that the crystal is coated 
with a layer of H or OH ions (Fig. 5.13). This charge development is similar to that 
developed by silicates. The sesquioxides have no permanent charge so their charge 
depends solely on, and varies greatly with, the pH of the soil solution. Allophane, 
an amorphous hydrous oxide with high surface area, also develops pH-dependent 
charge. Because its surface area is greater than crystalline materials, its charge is 
even more pH dependent. 

Soil organic matter also has a strongly pH-dependent charge. The charge develops 
mostly by H + dissociation from carboxylic and phenolic groups. Table 5.5 summa- 
rizes the colloidal properties of the major components of the soil's clay fraction. 



r 



1/2 



0-I/2 | /O 
\ ^-1/2 



-1/2 



Fe + 



0< /2 |'l/>-o 



1 



O j interfoce 



CX^2 r'^-W+l/2) 



Fe 3+ 



-1/2 



O -"2^6H(-l/2) 

o 



FIGURE 5.13. Fe 3 ' and ligands in the interior and at the surface of hematite. 



C3 >-i 

o > 

u < 



0> 






V 



c3 ^ 
D. Xl 
X - 

w 



D. C 
1/5 w 



I 

« v °P 

4> <N 



1^, 



E 
01 

c 
o 
a 
£ 
o 
o 

TO 



o 

(fl 



(0 
O) 

1_ 

o 

c 



0) 



a. 
o 



at 



o 

E 
£ 

in 
iri 



4> -^ i 



u *G s 



^ 



w 



1- *> 

3d 



3 

£ 

t-. 



<u a, 



1- I i 



O . 



> 



2£ 






08 
2§ 



O 

8 

33 

— ■ o 
oo 






DC 
O 



CO 

5 s 



»n 



I 






ESS 



.1-8 « 



000 
Z Z Z 



3S2 



000 

o N in 
00 — — 

© r- t— 



c 



o 

o 



o o 
o o 



8 

•o _ _ 

332 

000 

<N <N CN 



i 



o — ' 



£U —r 1 



«< 



CO 



■ DC 



X 

o 



o 



CO 



50 



< Z Z 5»2 



« .* < 5 .J? 
2i £. co 



3 I 



•C 3 
O 3 

E .2 



'E 

B o -2 
>P >5 -^ 
> S U 



a. 
o 



150 



SURFACE AREA MEASUREMENTS 151 

APPENDIX 5.1 SURFACE AREA MEASUREMENTS 

An impressive property of colloids, including layer silicate minerals, is their large 
area of reactive surface. Various physical and chemical properties, including water 
retention and cation exchange capacity, are highly correlated with the surface area of 
soils. Several techniques estimate the amounts of reactive surface area of soils and 
are briefly described below. 

Colloid chemists commonly measure surface area by the adsorption of N2 gas. 
The adsorption is conducted in vacuum and at temperatures near the boiling point of 
liquid nitrogen (—196° C). The approach is based on the Brunauer-Emmett-Teller 
(BET) adsorption equation, and has been adapted to a commercially available instru- 
ment. Unfortunately, the technique does not give reliable values for expansible soil 
colloids such as vermiculite or montmorillonite. Nonpolar N2 molecules penetrate 
little of the interlayer regions between adjacent mineral platelets of expansible layer 
silicates where 80 to 90% of the total surface area is located. Several workers have 
used a similar approach with polar H2O vapor and have reported complete saturation 
of both internal (interlayer) and external surfaces. The approach, however, has not 
been popular as an experimental technique. 

Soil chemists more commonly measure the retention of polar liquids such as ethy- 
lene glycol or glycerol by soils. The basic procedure involves applying excess and 
then removing all but a monolayer from the mineral surfaces. The excess is removed 
under vacuum in the presence of a desiccant, to eliminate competition with H2O 
for retention sites. Some workers advocate a glycol-CaCb mixture to maintain a 
relatively constant vapor pressure of glycol in the evacuated system, and hence to 
provide a more reproducible endpoint. 

Glycol and glycerol retention are influenced by the species of exchangeable 
cation, since both the colloid surfaces and the surface cations are at least partially 
solvated during surface area determinations. Glycerol is preferred over glycol by 
some workers, because it distinguishes between vermiculitic (partially expanding) 
and montmorillonitic (freely expanding) surfaces under carefully controlled condi- 
tions. A single molecular layer of glycerol remains in vermiculitic interlayers, but 
two such layers remain in montmorillonitic interlayers. 

Ethylene glycol monoethyl ether (EGME) is another polar molecule used increas- 
ingly for surface area measurements. Its results are essentially identical to the glycol 
method but are achieved more rapidly. It was graciously contributed to soil chem- 
istry by a careless shipping clerk and an unknowing technician. The latter obtained 
unusual, but promising, results before he realized that the wrong reagent had been 
provided by a chemical supply firm. 

Surface areas have also been measured by anion repulsion or by adsorption of cer- 
tain organic solutes from aqueous solution. A particularly promising solute is cetyl 
pyridinium bromide, which orients differently on external and internal (interlayer) 
surfaces and can thus aid in distinguishing between the two types of surface. 



1 52 INORGANIC SOLID PHASE 

APPENDIX 5.2 MINERAL IDENTIFICATION IN SOILS 

X-ray diffraction has probably contributed more to the mineralogical characteriza- 
tion of soil layer silicates than any other single technique. Other techniques being 
increasingly used are infrared, electron spin resonance, fluorescence spectroscopy, 
differential thermal analysis, and x-ray absorption spectroscopy. The simplest 
and most common is x-ray diffraction, which exposes material to a filtered and 
monochromatic beam of x-rays from an appropriate metal target. When the beam 
enters the sample, part of the beam is reflected by successive repeating planes of 
atoms. The reflected beams are reinforced (intensified) at each locus of points where 
the reflected beam has moved an integral number of wavelengths before being re- 
flected by the next plane of atoms (Figure 5 . 1 4). In quantitative terms, reinforcement 
occurs wherever 

nX = 2ds\n0 (5.2) 

where n is an integer, X is the wavelength of the x-radiation, d is the "repeat" dis- 
tance between successive layers of the crystal, and 6 is the angle at which the radi- 
ation strikes the crystal. The loci of points can be detected either with a cylindrical 
photographic film placed around the irradiated sample or with a rotating detector. 
Radiation that has not traveled an integral number of wavelengths within the crystal 
emerges and strikes the film or detector out of phase with other radiation, so only 
minimal film darkening or detector counts are recorded. Radiation that has traveled 
an integral number of wavelengths within the crystal reinforces previously reflected 
radiation and produces strong film darkening or a peak of counts in the detector. 
Differences in crystal repeat distances as small as 0.01 to 0.001 nm can be detected 
by x-ray diffraction. The technique is particularly valuable for identifying soil col- 
loid types, their degree of interleafing or interstratification, and variations in their 
interplatelet spacings resulting from pretreatments or additives. 

All the techniques to identify soil minerals have difficulty coping with the het- 
erogeneity of soils and with coatings of organic and weathered materials on soil 
particles, and have trouble detecting small amounts of a component in a very large 



FIGURE 5.14. X-ray reflection from repeating mineral planes. For reinforcement, CP+ PE ■■ 
2d sin = »//., so that emerging radiation is in phase. 



QUESTIONS AND PROBLEMS 1 53 

matrix of silicates and Fe and Al hydroxyoxides. If one constituent of the component 
has a much higher atomic weight, it can usually be more readily detected. Even Pb 
with atomic weight — 207, however, had to be <4% by mass before distinct Pb 
minerals could be identified in soil. A soil with that Pb content is a rich Pb ore. At 
normal Pb concentrations in soils, Pb and other trace elements more likely exist as 
isomorphous substitutes in the soil's major minerals. By going to great extremes to 
clean the mineral surface of coatings to reveal the minerals beneath, we may destroy 
the soil. 



BIBLIOGRAPHY 

Feklman, S. B., and L. W. Zelasny. 1998. In Chemistry of Soil Minerals (P. M. Huang, ed.). 

Soil Science Society of America Spec. Publ. 55, Madison, WI. 
Putnis, A. 1992. Introduction to Mineral Sciences, Cambridge University Press, New York. 



QUESTIONS AND PROBLEMS 

1. Distinguish between primary and secondary minerals, and give examples of 
each. Which minerals are more important in determining soil properties? 

2. Which minerals are commonly found in the sand and silt fractions of soil? Which 
are commonly found in the clay-sized fraction? Why? 

3. Distinguish between ionic, covalent, hydrogen, and van der Waals bonding. 
Which type of bonding predominates in silicate structures? 

4. What ion dominates silicate structures? 

5. Calculate the theoretical range in hole size between oxygen ions in tetrahedral 
and octahedral coordination. Which cations "fit" in each configuration? 

6. What is the dominant characteristic that determines whether ions may isomor- 
phically substitute for one another? 

7. What is a unit cell? How many unit cells are there in 1 mole of a particular 
mineral? 

8. Why is the phrase "clay mineral" misleading, and what term is best used to 
describe phyllosilicate minerals of >2 fim effective diameter? 

9. Distinguish between 1:1, 2:1, and 2:1:1 layer silicates by drawing diagram of 
their structures. 

10. Explain how layer charge influences the following layer silicate properties: 

(a) Interlayer bonds 

(b) Crystal size 

(c) Swelling 



1 54 INORGANIC SOLID PHASE 

(d) Surface area 

(e) c spacing 

(f) Exchangeability of adsorbed cations 

11. How do soil layer silicates differ from pure minerals? How are soil properties, 
such as cation exchange capacity, surface area, and swelling, expected to differ 
from the properties of pure layer silicates? 

12. Write the half-unit-cell formula for a montmorillonite mineral with a layer 
charge of 0.3, with 90° of the substitution in the octahedral layer (Mg 2+ for 
Al 3+ ) and 10% of the substitution in the tetrahedral layer (Al 3+ for Si 4+ ). The 
saturating cation is Na + . Draw a diagram of this mineral, indicating the position 
of Mg 2 +, Al 3+ , Si 4+ , and Na + . 

13. Calculate the cation exchange capacity of the mineral in Problem 12 and ex- 
press the result in millimoles(+) per kilogram. Compare this value to the value 
normally given for montmorillonite. 

14. The a and b dimensions of a typical montmorillonite mineral are a = 0.052 nm, 
and b = 0.089 nm. Using the data in Problems 12 and 13, calculate the theoret- 
ical surface area of the montmorillonite. (Remember that montmorillonite has 
both internal and external surfaces.) How does this value compare with values 
normally given for montmorillonite? 

15. Accessory minerals in the clay-sized fraction often dominate the properties of 
the soil solid phase. Under what conditions would you expect this to be true, 
what minerals are involved, and how would the soil's properties be affected? 

16. What are the likely dominant sources of pH-dependent charge in 

(a) A highly weathered mineral soil low in organic matter? 

(b) A slightly weathered, montmorillonitic soil low in organic matter? 

(c) A slightly acid forest soil? 

(d) A volcanically derived soil low in organic matter? 

17. What is meant by the zero point of charge? 

18. Hydrous oxides are said to be amphoteric. Explain. 

19. What are the functional groups responsible for pH-dependent charge in soils? 

20. Calculate the proportions of mass and volume occupied by oxygen in kaolinite 
and goethite. 

21. Would a hypothetical mineral of composition (Al2Mg)(Si3Al)Oio(OH)2 be 
dioctahedral or trioctahedral? 



6 



SOIL ORGANIC MATTER 



The organic fraction of soil is <5% living microbes, plant roots, and soil fauna and 
>95% dead plant and animal residues. Per unit mass, the organic fraction is the most 
chemically active portion of the soil. It is a reservoir for essential elements, partic- 
ularly C, N, S, and P; promotes good soil structure; is a source of cation exchange 
capacity (CEC) and soil pH buffering; promotes good air-water relations in soils; 
and is a large and active reservoir and buffer of carbon in the environment. This 
chapter discusses the contributions of soil organic matter to the chemical properties 
of soils. 

Soil organic matter (SOM) is mostly (>95%) an accumulation of dead plant mat- 
ter and partially decayed and resynthesized plant and animal residues. Freshly fallen 
leaves and dying roots rapidly decompose and the residues become part of SOM, 
some portions of which remain in the soil for a very long time. Crop residues, weeds, 
grasses, tree leaves, worms, bacteria, fungi, and actinomycetes are also part of the 
complex mixture. Some definitions of SOM are restricted to soil humus, omitting any 
undecayed organic residues and soil organisms. Humus is generally defined as that 
organic material that has been transformed into relatively stable form by soil microor- 
ganisms. We use the term SOM in its broader sense, to include all carbon-containing 
compounds except carbonates. Stevenson (1982) and Schnitzer and Schulten (1998) 
give details on the details of SOM composition and structure. This chapter is con- 
cerned with nonliving substances (particularly humus) and its chemistry. Soil micro- 
biology texts discuss the live microbial tissue {microbial biomass) in soils. 

6.1 SOIL ORGANIC MATTER CONTENT 

Soils vary greatly in their organic matter content. A prairie grassland soil (e.g., Mol- 
lisol) may contain 5 to 6% SOM by mass to a depth of 1 5 cm, whereas a sandy desert 

155 



156 SOIL ORGANIC MATTER 

soil (Aridisol) contains little more than 0.1%. Poorly drained soils (Aquepts) often 
have organic matter contents greater than 10%, and peat soils (Histosols) approach 
100% organic matter. Although the organic matter content of most surface mineral 
soils is only 0.5 to 5% by mass, the active colloidal behavior of SOM strongly affects 
soil physical and chemical properties. 

The factors of soil formation (Section 7.3) determine the SOM content of soils. 
The order of importance of the factors that determine the organic matter (and nitro- 
gen) contents of well-drained soils in the United States is climate > vegetation > 
topography = parent material > age. 

Climate affects (1) the array of plant species, (2) the quantity of plant material 
produced, and (3) the intensity of soil microbial activity. Vegetation and topographic 
effects are difficult to separate from climatic effects. Rather, all of the factors become 
integrated as a soil forms and account for the generalization that forest and grassland 
soils usually exceed other well-aerated soils in humus content, whereas desert and 
semidesert soils have very little SOM. 

Tropical soils, both humid and arid, were once thought to have low organic matter 
contents because the soils lack the dark color that characterizes SOM in temperate 
regions, and because temperature is inversely correlated with SOM content in tem- 
perate soils. The high soil temperature should increase microbial oxidation of SOM. 
The organic matter contents of similar soil orders in tropical and temperate regions, 
however, are quite similar (Table 6.1). The high organic matter contents of the tropi- 
cal soils may be due to strong interaction of organic matter with Fe and Al hydroxy- 
oxides and allophane. Strong interaction could stabilize the SOM against microbial 
decay. The high SOM content is also due to the high rate of year-round biomass 
production in the humid tropics. 

Topography influences the amount of SOM in two ways. North-facing slopes (in 
the northern hemisphere) are cooler and moister, so the organic matter content is 
greater than in soils on south-facing slopes. A much greater effect is topography's 
control of water drainage. In poorly drained and swampy soils, plant matter can 
be covered with water, and oxygen excluded, as soon as the plant dies. In stagnant 
water, oxygen diffusion from the surface is the only means of oxygen supply to the 



Table 6.1. Average organic carbon contents of several soil orders in temperate and 
tropical regions (to obtain approximate SOM values, multiply by 1.7) a 





depth) 




Carbon Contents (%) 




Soil Order 
(Oto 15 cm 


United 
States 


Brazil 


Zaire 


Mean 

Level 


Mollisols 
Oxisols 
Ultisols 
Alfisols 




2.44 

1.58 
1.55 


2.01 
1.61 
1.06 


2.53 

0.8 

1.30 


2.44 
2.07 
1.39 
1.30 



" I'iom P. A. Sanchez. 1 976. Properties and Management of Soils in lite Tmpics. Wiley-Inlerscience, New 
York. 



SOIL ORGANIC MATTER CONTENT 1 57 

organisms carrying out the decomposition. This slow rate of oxygen supply preserves 
much of the plant matter from decay. Fermentation can operate in the absence of 
oxygen, but its effect on decay is rather small. Running water can bring enough 
dissolved oxygen to oxidize organic matter in streams. The turnover of lake water in 
spring and fall supplies some oxygen to lake bottoms. 

In stagnant water, dead and partially decayed plant matter can accumulate to the 
water surface. Such deposits were up to 30 m thick in the Sacramento-San Joaquin 
Delta of California, for example, before those areas were drained. Ireland has large 
areas of peat (a type of Histosol containing 50% or more by mass of organic matter) 
deposits more than 10 m thick. Low temperatures also enhance peat accumulation. 
Major areas of peat are found in central Canada, which is both cold and swampy, with 
lesser amounts in low-lying areas of northern Russia and northern Europe. Smaller, 
but substantial, areas of Histosols are found on every continent and even in the trop- 
ics. The largest area of contiguous peat or muck soils in the continental United States 
is in the Everglades area south of Lake Okeechobee in southern Florida. 

Peat accumulation depends on the rate of organic addition versus the rate of oxi- 
dation. The oxidation rate is a function of the rate of oxygen supply and temperature. 
The oxygen supply in turn depends on soil water content. When drained for cultiva- 
tion, peat oxidizes as it contacts oxygen and also shrinks as it dries. Shrinkage and 
oxidation lowered the level of cultivated peat land in California and Florida by as 
much as 25 to 50 mm yr ' . This rate has slowed in recent years as we have learned 
to better manage these materials. The peat islands in California are encircled by 
dikes to keep out the water. The peat surface is now as much as 10 m below the river 
level. The dikes must be continually strengthened as the peat oxidizes and shrinks. 
In recent years the maintenance of these dikes has become so costly that some of 
the islands have been abandoned after winter floods broke through the dikes. Wind 
erosion of the dry surface and accidental burning of the peat also contributed to the 
loss of these extremely productive soils. 

In the formerly glaciated areas of Canada, Europe, and Siberia, peat has been 
accumulating for 10000 to 15000 years, since the glaciers retreated. This accumu- 
lation rate varies from year to year, with a rough average of about 0.1 mm yr -1 in 
northern Canada. Since these areas were scraped bare by the glaciers, they represent 
a great transfer and redistribution of SOM and biomass carbon from more southern 
regions, and of CO2 from the atmosphere, to the peat lands. The coal and petroleum 
deposits of the Carboniferous Era formed from enormous accumulations of peat soils 
in many areas of the world. They represent beds of SOM that were laid down under 
conditions similar to those that produce the peat lands of today. 

Parent material influences SOM contents mainly through its effect on soil texture. 
In an area of similar climate and topography, SOM content tends to increase with soil 
clay content. The intimate association of humic substances with inorganic solids as 
organomineral complexes preserves organic matter. Montmorillonitic clays because 
of their high surface areas have particularly high adsorptive capacities for organic 
molecules and are particularly effective in protecting nitrogenous constituents from 
attack by microorganisms. This strong interaction between clays and organic matter 
also gives rise to important effects of SOM on soil physical and chemical properties. 



1 58 SOIL ORGANIC MATTER 

As conditions change so does the SOM content. If humanity or nature lays soil 
bare, the SOM content can recover rapidly by geologic time standards to its previous 
state, but not fast enough for the human inhabitants of the area. Several years after 
the volcanic eruptions of Mt. St. Helens in Washington and of Mt. Pinatubo in the 
Phillipines, plants are reestablishing themselves on fresh-deposited pumice. Mine 
spoils and their revegetation also show this behavior. Plant growth is sparse in the 
first years as sulfides in the ores oxidize to sulfuric acid and as the rock minerals 
begin to weather. After the acid is leached out and some weathering has occurred, 
the plant increase and the rate of SOM increase follow an S-shaped curve. They start 
slowly, then increase exponentially, and the rate eventually levels off to a steady-state 
SOM content determined by climate and topography. The steady state may require 50 
years for fine-textured parent material in humid climates, or as much as I500 years 
for sandy soils in arid regions. These "new" soils also include flood plain deposits, 
road cuts, and land no longer cultivated. 

Farming more subtly affects SOM. SOM levels may decrease by as much as one- 
half when unfilled lands are cleared and cultivated. The oxidation rate is relatively 
fast during the first years as the cultivating tools shear soil aggregates and expose 
SOM that had been relatively protected from microbial attack by a coating of more 
resistant organic matter. The "minimum tillage" form of cultivation has the goals of 
minimizing the energy costs of cultivation and of increasing the SOM and nitrogen 
content. 



6.2 THE DECAY PROCESS 

The decay of SOM is the oxidation of organic carbon by heterotrophic organisms 
that utilize the energy of oxidation for their metabolism. The initial breakup of tree 
trunks and large objects is carried out by animals foraging for grubs, by termites, and 
by earthworms. Saprophytic plants such as mushrooms and snowfiowers also obtain 
their energy from this partially decomposed plant matter. As the organic matter be- 
comes more finely divided, the size of the decomposing organisms also decreases. 
Decay proceeds as long as the oxygen, water, temperature, and nutrient levels are 
adequate for the decomposing organisms. In the desert, the absence of water greatly 
hinders the oxidation rate of organic material at the soil surface. Beneath the surface 
of arid soils, where the moisture content is more likely to be adequate, decomposition 
is rapid. 

The annual input rate of dead plant matter to soils in temperate regions, the net 
primary productivity of plants, is about 1 kg m~ 2 of carbon or 20 Mg ha - ' of dry 
matter. The annual input rate is perhaps 2 kg C m -2 or 40 Mg ha -1 of dry matter in 
humid tropical forests, and decreases to virtually zero to 0. 1 kg C m -2 in deserts and 
arctic tundra. This input is part of the nonhumus organic matter, which includes orig- 
inal plant and microbial tissue and partially decomposed material. These nonhumus 
substances contain carbohydrates and related compounds, proteins and their deriva- 
tives, fats, lignins, tannins, and various partially decomposed products in roots and 
plant tops. The portion contributed by dead animal matter is insignificant because the 



THE DECAY PROCESS 



159 



amount of carbon as plant biomass is 10000 times greater than the carbon in animal 
biomass. 

New plant material is an excellent source of food for soil microorganisms. Mi- 
crobes are selective: Simple monomers of sugars, amino acids, and fats are oxidized 
first. The polymers — starch and proteins — follow close behind. Cellulose, the most 
prevalent plant component, decomposes more slowly. Lignin and hydrocarbons de- 
compose even more slowly. Some plant materials (e.g., pine needles and oak leaves) 
and some synthetic organic chemicals contain inhibitory compounds and decompose 
slowly and by only a few microbial species. Shortly after fresh material contacts the 
soil, microbes begin decomposing it as a source of nutrients and energy (Fig. 6.1). 
The initial phase of microbial attack is a rapid loss of easily decomposable organic 
substances. Molds and spore-forming bacteria are especially active in consuming 
proteins, starches, and cellulose. Their major products are CO2 and H2O plus a small 
amount of new microbial tissue. By-products, especially in partially anaerobic con- 
ditions, include small amounts of NH3, H2S, organic acids, plus other incompletely 
oxidized substances. In subsequent phases of decomposition, when aerobic condi- 
tions are present, these intermediate compounds and newly formed microbial tissues 
are attacked by other microorganisms, with production of some new biomass while 
a larger fraction of the organic carbon is converted to CO2. 



Plant and animal residues 



-CO, 



H 2 0, NH 3 , 
H,S, etc. 



Cellulose 
hemicellulose, 
proteins, etc. 



Microbial 
decomposition 



Microbial protoplasm 

and metabolic 

byproducts 



Aromatic 
structures 



Lignins 
tannins, etc. 



Microbial 
modification 



Modified lignins 
and aromatics 



CO, 

h 
H 2 0, NH 3 ,etc. 



Amino acids, 
proteins, etc. 



Humus 
. mineralization - 



FIGURE 6.1. Organic matter decomposition and formation of humic substances. (From 
F. E. Bear, ed. 1964. Chemistry of the Soil, ACS Monograph Series No. 160, p. 258.) 



160 



SOIL ORGANIC MATTER 



100 



* 75 

a 

z 

I SO 



IU 

c 



25 - 





l 


I i 
1. Total C Remaining 


-\\ 




2. Raalstant Plant C 

3. Microbial Product* 


. \ \ 


\1 




\2 




1 1 


i 



20 40 

TIME In DAYS 



80 



FIGURE 6.2. Idealized diagram for the decay of crop residues in soil under conditions that are 
optimal for microbial activity. (From F. J. Stevenson. 1982. Humus Chemistry. Wiley, New York.) 



Figure 6.2 is an idealized graph of crop-residue decay with time in temperate re- 
gions. Plant residues are attacked rapidly at first, but the rate of decay soon slows. 
Considerable plant carbon remains in the soil at this point, but part of the residual 
carbon occurs as microbial by-products and part as the more resistant plant residues. 
Different plant components decompose at different rates. Lignin in wood is attacked 
much more slowly than cellulose. While most is converted into CO2, the more read- 
ily decomposable constituents are also partially resynthesized by the decomposer 
microbes into more resistant components. Decay also slows because microbes work 
on the surface of particles; the resistant material can coat the underlying material and 
protect it from further decay. 

Figure 6.3 shows the course of carbon losses from soils in a temperate climate 
(England) and a humid tropical climate (Nigeria). The shapes of the curves are sim- 
ilar, but the time scale is four times faster in the tropical climate. The approximate 
half-life of fresh organic matter in temperate regions is about 3-4 months, but it is 
as little as 3-4 weeks in the humid tropics. The flatter portion of the cube represents 
the second stage of organic matter decomposition, with a half-life of about 1.6 years 
in the humid tropics and 6.2 years in temperate regions. 

The final stage of decomposition is the gradual loss of the more resistant plant 
parts, such as lignin, in which the actinomycetes and fungi play a major role. A 
small fraction of the original carbon, however, persists for a very long time. When 
the age of soil organic carbon is measured by 14 C dating, a small but extremely old 
fraction raises the average age of carbon in SOM to about 1000 years in surface soils 
and to several thousand years in subsoils. 

Results of field experiments confirm that carbon becomes increasingly resistant 
to decomposition with time. This has led some investigators to conclude that the 
organic component exists in three major fractions when considered on a dynamic 
basis: (1) decomposing plant residues and the associated biomass, which turn over 
every few years; (2) microbial metabolites and cell wall constituents that become 
stabilized in soil and have a half-life of 5 to 25 years; and (3) the smallest fraction, 
resistant organic matter, ranging in age from 250 to 2500 years or more. 



EXTRACTION, FRACTIONATION, AND COMPOSITION 



161 



PERIOD OF INCUBATION IN ENGLAND v«»r« 




1.6 tO 2£ 

PERIOD OF INCUBATION IN NIGERIA years 



FIGURE 6.3. Decomposition rates of fresh organic matter in England and in Nigeria. (From 
D. S. Jenkinson and A. Ayanaba. 1970. Soil Sci. Soc. Am. Proa 43:912.) 



Despite the stability of the resistant fractions of SOM, 50 to 80% of freshly added 
organic matter is lost from most temperate soils during the first year. The smaller the 
particle size, the faster the SOM is destroyed. Plant or animal residues must be added 
to soils continually if the favorable effects of organic matter on soil properties are to 
be maintained. To increase the SOM content by adding organic residues to soils is 
difficult. The decomposition rate of organic materials in soils increases in proportion 
to the addition rate. The more organic matter added, the more it is oxidized. An 
experiment in Rothamsted, England, has been adding ca. 30 tonnes ha - ' of organic 
manures to a cultivated soil annually since 1843, without bringing the SOM content 
back to its uncultivated level. 



6.3 EXTRACTION, FRACTIONATION, AND COMPOSITION 



Soil organic matter is a complex polymer that can be studied only after it is separated 
from the inorganic soil fraction and after it has been broken into smaller fragments. 
Recent advances have lessened the degree to which SOM has to be destroyed to be 
studied. The separation of organic matter from the inorganic matrix of sand, silt, 
and clay is not physically difficult, but the extracting agent (traditionally O.l to 0.5 
M NaOH) is harsh and alters the organic matter through hydrolysis and autoxida- 
tion. The components in such extracts can be partially fractionated by precipitation 
with acids or metal salts, or by taking advantage of solubility differences in various 
organic solvents. Students interested in the extraction and fractionation procedures 
should consult the works of Stevenson (1 982) and Schnitzer and Schulten (1998). 



1 62 SOIL ORGANIC MATTER 

The classical procedure for fractionation of extracted organic matter involves acid 
precipitation of some fractions from an NaOH extract, and subsequent dissolution of 
part of the precipitated material with alcohol (Fig. 6.4). The humic acids and fulvic 
acids fractions so prepared are mixture of many different chemical compounds in 
various stages of polymerization. Stevenson defines them as humic acid, the dark- 
colored organic material that is extracted from soil by various reagents and that pre- 
cipitates when dilute acid is added, and fulvic acid, the colored material that remains 
dissolved in the extracting solution after acidification. Schnitzer and Schulten define 
humic acid as the fraction soluble in dilute alkali or neutral salt but coagulated by 
acidification, and fulvic acid as that which remains in aqueous solution after acidifi- 
cation, and humin as the fraction that cannot be extracted from soils by dilute base, 
neutral salts, or acids. Humin is thought to be humic acid-like material that has re- 
acted strongly with inorganic soil components and thereby resists attack by alkalis 
and acids. 







Soil organic matter 




I 

Strongly colored 
humic matter 

I 
Treat with alkali 

I 






Nonhumic matter 

(undecomposed plant 

remains, etc.) 


I 

Insoluble 

humin 


I 

Soluble fraction 

I 
Treat with acid 

I 




I 

Not precipitated 
Fulvic acid 




I 
Precipitated 
Humic acid 





Redissolve in base and 
add electrolyte 



Not precipitated 
Brown humic acic 

Gray humic acid alcohol 



. • _. Bro wn humic acid ,-,... -.u 

Precipitated Extract with 



Hymatomelanic acid 

FIGURE 6.4. Fractionation of soil organic matter. (Modified from F. J. Stevenson. 1982. Humus 
Chemistry. Wiley, New York; and J. L. Mortenson and F. L. Hines. 1964. In Chemistry of the Soil 
(F. E. Bear, ed.). ACS Monograph Series No. 160.) 



EXTRACTION, FRACTIONATION, AND COMPOSITION 



163 



Organic materials undergo microbial enzymatic and chemical reactions in soils 
to form colloidal polymers called humus (Fig. 6.1). Humus is a complex and rather 
microbially resistant mixture of brown to almost black, amorphous and colloidal 
substances modified from the original plant tissues or resynthesized by soil organ- 
isms. Humus contains approximately 10% carbohydrates, 10% nitrogen components 
(proteins, amino acids, and cyclical N compounds), 10% "lipids" (including alkanes, 
alkenes, fatty acids, and esters), and 70% humic substances. 

Chemists have attempted to unravel the details of humus composition for many 
years. Despite considerable progress in characterizing various extracts, much re- 
mains to be discovered. Humus contains primarily C, H, O, N, P, and S plus small 
amounts of other elements. Only a small but important portion is soluble in water, but 
much is soluble in strong bases. The various fractions of humus obtained on the basis 
of solubility characteristics are part of a heterogeneous mixture of molecules, which 
range in molecular weight from as low as several hundred to over 300 000 (Fig. 6.5). 
Carbon and oxygen content, acidity, and degree of polymerization all change system- 
atically with molecular weight. The low molecular weight fulvic acids have higher 
oxygen contents and lower carbon contents than the high molecular weight humic 
acids. The more soluble fulvic acids are usually responsible for the brownish-yellow 
color of many natural waters. Humic acids precipitate with acids and polyvalent 
cations, thus tending to be insoluble. 

Recently, SOM has begun to be analyzed by pyrolysis, which is destruction at high 
temperatures in the absence of oxygen, and analysis of the many volatile compounds 
that emanate from SOM. Carbohydrates, phenols, lignin and n-fatty acids are the 



(Decomposition products of organic residue*) 




LNonfcumAc •ubetaftC.ee, I 
(Known classes of organic compound*) 



(Pigmented polymers) 




Crenic acid Apocrenlc tcld 

(Berzellus) 



Brown humic acids Gray humic acids 
(Spring*!") 



Light yellow 



Yellow brown 



Gray-black 



„ Increase In degree of polymerization- 

2 000? Increase In molecular weight 

45% increase In carbon content - — 

49%- decrease In oxygen content ------ 

1,400 decrease in exchange acidity 



-300 000? 

. _a»-82% 

— ■*.30% 

— *-500 



FIGURE 6,5. Classification and chemical properties of humic substances. (From F. J. Steven- 
son. 1982. Humus Chemistry. Wiley, New York.) 



164 SOIL ORGANIC MATTER 

major products volatilized. Pyrolysis is a destructive technique but the compounds 
evolved may lead to a belter understanding of humus and soil organic matter. 

The composition and structure of soil humus probably varies with the material 
from which it was derived and with the species of microbe. Although there is no rea- 
son to believe that humus from different soils is the same, several researchers agree 
on a "type" structure for humic acid (Fig. 6.6). Two types of polymers, humic acid 
(50 to 80% by mass) and polysaccharides (10 to 30% by mass), can constitute up 
to 90% or more of the total humus in soils. A typical humic acid molecule probably 
consists of polymers of a basic six-carbon aromatic ring structure of di- or trihy- 
droxyl phenols linked by — O — , — NH — , — N — , and — S — bonds, and containing 
— OH groups and quinone ( — O — C6H4 — O — ) linkages. This structure contains a 
high density of reactive functional groups. Individual humic acid molecules vary in 
the structure and density of functional groups, but the basic structure is thought to 
remain approximately the same. 

The chemical origin of humus components is not yet resolved. Some workers con- 
sider humus to be microbially resistant plant materials (lignin, suberin, cutins, paraf- 
fins, etc.) that are awaiting oxidation to humic acid and further oxidation to fulvic 
acid. Some have thought that humic materials were the hugely nondegraded plant 
materials, which would account for the aromatic content; microbiologists thought 
that producing large amounts of aromatic groups was too exotic for soil microbes. A 
lignin origin for humus, however, would not account for the nitrogen content of hu- 
mus or for the large amount of alkanes (paraffins) found in recent studies. Lignin in 
wood is nitrogen-free and synthesis of alkanes (aliphatic hydrocarbons) was thought 
to be rare. The aromatic (alkene) components have received hugh attention: They 
are present in lignin and make up about half of the carbon in SOM. Recent studies 
indicate that alkanes are also present in SOM and may play a role in binding the 
aromatic groups together. 



6.4 COLLOIDAL PROPERTIES 

Most of the colloidal properties of SOM are due to humus. Humus is highly colloidal 
and is x-ray amorphous rather than crystalline. The surface area and adsorptive ca- 
pacities of humus per unit mass arc greater than those of the layer silicate minerals. 
The specific surface of well-developed humus may be as high as 900 x 10 3 m 2 kg -1 ; 
its exchange capacity ranges from 1500 to 3000 mmol(+) kg -1 . 

The negative charge (and hence the CEC) of humus is generally agreed to be due 
to the dissociation of H + from functional groups. All charge on humus is strongly 
pH dependent, with humic and fulvic acids behaving like weak-acid polyelectrolytes 
(polyprotonated weak acids). Figure 6.7 shows typical titration curves for peat and 
soil humic acids. Both buffer soil pH over a wide range. The slopes of humic and 
fulvic acid titration curves do not change as sharply as for monomeric acids, be- 
cause of electrostatic effects on the high molecular weight polyacids and because of 
configurational changes that occur at higher pH. 




CD 

.C 



c 

.0) 

c 
o 






O 



cr 



a> 



e 
p 

LL 



^ 

1 5 



•ft o 

o ,co 

£ CO 



/A Vr 



1< 
■5 O 

°- ■>. 

>, 2? 

x ,9> 
o 

3* 

UJ ro 
£E .9 

<2 £ 
u. < 



165 



166 



SOIL ORGANIC MATTER 

10 



8 - 



pH 



6 - 



4 - 





"" i ■ i ' i 

A. PEAT HUMIC ACID 


I 


I ' 




y 




B. SOIL HUMIC ACID 






/ / 




- 






1 I 


f\/\ 


— 


- 






/ 1 f* 


- 


- 


— 


I y 




/■■ 




— 




A yx^ 


"B 








r 


I I I I I 


I 


i i 


I i 


r 



.04 



.08 



.12 



.16 



.20 



.24 



BASE . mmole 



FIGURE 6.7. Titration curves of a soil and peat humic acid. The small wavy lines on the curves 
indicate endpoints for ionization of weak-acid groups having different, but overlapping, ionization 
constants. (From F. J. Stevenson. 1982. Humus Chemistry. Wiley, New York.) 



The dissociation of carboxyl and phenol groups yields perhaps 85 to 90% of the 
negative charge of humus. Many carboxylic groups are sufficiently acid to dissociate 
below pH 6 (zone I, Fig. 6.7) 



R— COOH = R— COO - + H+ 



(6.1) 



leaving a negative charge on the functional group. Here R represents any number 
of organic species whose differing electronegativities alter the tendency for H + to 
dissociate. Thus, the various R — COOH units dissociate at different pH values. As 
the pH of a system increases above 6, still weaker carboxylic groups and other very 
weak acids dissociate (zone II, Fig. 6.7). Zone III represents dissociation of phenolic 
OH and other very weak acids at pH > 8. Dissociation of H + from acid groups 
throughout the pH range adds to the total negative charge of humus. Dissociation of 
H + from enolic OH, imide (=NH), and possibly other groups also contribute to the 
negative charge. 

No SOM fraction possessing a net positive charge has been found at normal soil 
pH values (Table 6.2). Protonated groups such as (R— OH 2 ) + and (R— NH^)" 1 " can 
yield positive charges, but the overall charge on humus is negative. 

Charged sites (primarily COO~) enable SOM to retain cations in nonleachable but 
exchangeable forms that are available to plants. The bonding is primarily coulombic 



FUNCTION OF ORGANIC MATTER IN SOIL 1 67 

Table 6.2. Effects of pH on cation exchange capacity (CEC) for 
60 Wisconsin soils" 





CEC (mmol(-t-) kg -1 , Means for 60 Soils) 












%CEC 




Organic 


Layer 


Whole 


from 


PH 


Matter 


Silicates 


Soil 


SOM 


2.5 


360 


380 


58 


1 


3.5 


730 


460 


75 


28 


5.0 


1270 


540 


7 


37 


6.0 


1310 


560 


108 


36 


7.0 


1630 


600 


123 


40 


8.0 


2130 


640 


148 


45 



"Data of C. S. Helling, G. Chesters, and R. B. Corey. 1964. Soil Sci. Soc. 
Am. Proc. 28:517-520. 



or electrostatic (e.g., — COO K + ). The bonding is also partly covalent, particularly 
when the charge is neutralizxd by transition-metal cations (Fe 2+ , Zn 2+ , Cu 2+ , Ni 2 '"). 



6.5 FUNCTION OF ORGANIC MATTER IN SOIL 

Table 6.3 summarizes the general properties of humus and its associated effects on 
soil. SOM contributes to plant growth through its effects on the chemical, biologi- 
cal, and physical properties of soil. The SOM supplies N, P, and S for plant growth, 
serves as an energy source for soil microorganisms, and promotes good soil struc- 
ture. Humus also indirectly affects the plant uptake of microelement and heavy metal 
cations, and the performance (availability) of herbicides and other agricultural chem- 
icals. SOM is highly porous so that pesticide and other organic compounds added to 
soils can be enveloped by SOM. In that form they are much less active. Soils with 
high amounts of SOM require higher dosages to herbicides and pesticides to achieve 
the desired result. 

The SOM supplies nearly all the N, 50 to 60% of the P, perhaps as much as 80% 
of the S, and a large part of the B and Mo adsorbed by plants from unfertilized, 
temperate region soils. Indirectly, SOM affects the supply of essential elements from 
other sources. The amount of N2 fixation by the free-living bacterium Azotobacter, 
for example, is related to the amount of readily available energy sources in the soil, 
such as the carbohydrates in SOM. 

In humid soils, C, N, and S are found predominantly in SOM. With increasing 
aridity, the amount of organic matter decreases and the fractions of the inorganic 
forms of the elements (carbonate, sulfate, and nitrate) tend to increase. 

The mass ratio of C/N/S in the SOM of temperate region soils is roughly 
100/10/i. Carbon supplies the energy for N and S reduction, as well as the matrix 
of compounds into which reduced N and S are incorporated and stabilized. Nitrogen 



168 



SOIL ORGANIC MATTER 



Table 6.3. General properties of humus and associated effects in soil 3 



Property 



Remarks 



Effect on Soil 



Color 



Water retention 



Combination with 
clay minerals 

Chelation 



Solubility in water 



pH relations 



Cation exchange 



Mineralization 



Combination with 
organic molecules 



The typical dark color of 
many soils is caused by 
organic matter 

Organic maueT can hoi (J up to 
20 times its weight in water 

Joins soil particles into 

structural units called 

aggregates 

Forms stable complexes with 

Cu 2 +.Mn 2+ ,Zn 2+ , 

and other polyvalent cations 

Insolubility of organic matter 

results partially from its 

association with clay; salts 

of divalent and trivalent 

cations with organic matter 

are also insoluble; isolated 

organic matter is partly soluble 

in water 

Organic matter buffers 

soil pH in the slightly 

acid, neutral, and alkaline 

ranges 

Total acidities of isolated 

fractions of humus range 

from 3000 to 14 000 mmole 

kg" 



-I 



Decomposition of organic 
mailer yields C0 2 , NHJ 



V>- 



,2- 



NO^, PO4 , and SOJ 
Affects bioactivily, persistence, 
and biodegradability of pesticides 



May facilitate warming 



Helps prevent drying and 

shrinking; improves moisture 

retention in sandy soils 

Permits gas exchange; 

stabilizes structure; increases 

permeability 

Buffers the availability of 

trace elements to higher 

plants 

Little organic matter is lost 

by leaching 



Helps to maintain a uniform 
reaction (pH) in the soil 



Increases the cation exchange 

capacity (CEC) of ihe soil; 

from 20 to 70% of the 

CEC of many soils is caused by 

organic matter 

A source of nutrient elements 

for plant growth 

Modifies the application rate of 
pesticides for effective control 



"From F. J. Stevenson. 1982. Humus Chemistry. Wiley, New York. 



and sulfur, in turn, are among the elements that govern the rate of plant growth and 
photosynthesis. The result under natural conditions is a relatively constant C/N/S 
ratio in SOM. 

The availability of many microelement cations is strongly affected by SOM. Vari- 
ous low molecular weight and somewhat water-soluble components of SOM, such as 
fulvic acid, form stable complexes (chelates) with Fe 2+ , Cu 2+ , Zn 2 '" and other poly- 
valent cations. These shield the cations from hydrolysis and precipitation reactions 



FUNCTION OF ORGANIC MATTER IN SOIL 1 69 




FIGURE 6,8. Postulated reaction between Cu 2+ and fulvic acid function groups. {After 
D. S. Gamble et al. 1970. Can. J. Chem. 48:3197.) 



and therefore increase their water solubility. A typical reaction between Cu 2+ and 
the functional groups of fulvic acid is given in Fig. 6.8. Inorganic precipitation, par- 
ticularly in soils of high pH, greatly reduces the solubility and availability of many 
of the essential microelements. Organic amendments (manure, sewage sludge, etc.) 
can improve microelement availability in alkaline soils and correct Fe and Zn defi- 
ciencies in particular. The amendments apparently release microelements in chelated 
form or release (mineralize) fulvic acid compounds that chelate the inorganic Fe and 
Zn present in the soil. SOM also combines with toxic ions such as Cd 2+ and Hg 2+ , 
as well as with microelement cations at high concentrations, and reduces their avail- 
ability. Organic amendments to soils often decrease cation toxicities in acid soils. 

Soil organic matter is involved with soil acidity. In mature Swedish forests, for 
example, the soil pH is perhaps 3.5. After harvest and while the succeeding trees 
are young, the pH is >4. The pH then decreases steadily as the forest matures, only 
to rise again when those trees are harvested. This pH cycle is thought to be due to 
organic acids formed by increasing litter fall. When litter production stops or slows 
and the organic acids decay or are leached away, the soil pH rises. In New England, 
part of the pH decline in formerly cultivated fields may be due to increased organic 
acids produced by greater leaf litter as trees invade the fields. In both cases, the pH 
is too low to be caused by Al 3 " 1 " hydrolysis, and Al hydrolysis would not account for 
the cyclical pH change. 

Humus affects soil structure and thus soil tilth, aeration, and moisture retention. 
The deterioration of structure that accompanies intensive tillage is usually less severe 
in soils adequately supplied with humus. When humus is lost, soils tend to become 
hard, compact, and cloddy. 

Aeration, water-holding capacity, and permeability are all improved by humus. 
The frequent addition of easily decomposable organic residues leads to the synthe- 
sis of complex organics (e.g., polysaccharides) that bind soil particles into aggre- 
gates. The intimate association of clay-sized particles (layer silicates) with humus 
via cation (e.g., calcium, magnesium, aluminum, iron) bridges also promotes aggre- 



170 SOIL ORGANIC MATTER 

gation. The water-insoluble salts of humic acid with polyvalent cations are called 
humates. They tend to be amorphous and glue-like. Heavy (clayey) soils, in partic- 
ular, benefit from organic matter additions by promoting particle aggregation. Ag- 
gregation yields a loose, open, granular structure for good water and air permeabil- 
ity- 
Humus also absorbs large quantities of water. The fully synthesized humus of a 

mineral soil contains as much as 80 to 90% water by weight. Additionally, micro- 
pores within larger soil aggregates hold available water for plants. This increase in 
plant-available water-holding capacity is a major benefit of organic matter additions 
to sandy soils. 

The data in Table 6.2 point out important characteristics of the CEC of SOM. All 
of the charge of humus is pH dependent, even at low pH. The CEC of both organic 
matter and layer silicates increases with increasing soil pH, but the CEC of SOM 
increases faster with pH than that of the layer silicates. In one soil, for example, the 
SOM contributed only l% of the total CEC at pH 2.5, but 45% at pH 8.0. In soils 
dominated by low-CEC layer silicates, such as kaolinite, the relative contribution of 
organic matter to whole-soil CEC can be even greater. A large fraction of the CEC 
in most soils is due to SOM. 

The functional groups responsible for the high CEC of humus also buffer soil pH 
over a wide range. This buffering contributes significantly to the lime requirement of 
acid soils (Chapter 8). Total acidities of isolated fractions of humus vary from 3 to 
14mol kg -1 . 



6.5.1 Organic Chemical Adsorption 

Organic matter content is the soil factor most directly related to the sorption of most 
herbicides and organic compounds by soils. The manner in which organic matter 
sorbs organics is discussed more fully in Chapter 7. The SOM content strongly influ- 
ences pesticide behavior in soil, including effectiveness against target species, phy- 
totoxicity to subsequent crops, leachability, volatility, and biodegradability. Recom- 
mended herbicide application rates are often higher for soils high in organic matter 
content, to compensate for greater adsorption in these soils. The soil behavior of 
organic chemicals, and particularly their interaction with SOM, is an active area of 
current soil chemistry research. 



BIBLIOGRAPHY 

Jenny, H. 1941. Factors of Soil Formation. McGraw-Hill, New York. 

Schnitzer, M., and H.-R. Schulten. 1998. New ideas on the chemical make-up of soil humic 
and fulvic acids. In Future Prospects for Soil Chemistry (P. M. Huang, ed.). Soil Science 
Society of America Spec. Publ. 55, Madison, WI. 

Stevenson, F. J. 1982. Humus Chemistry. Wiley, New York. 



QUESTIONS AND PROBLEMS 171 

QUESTIONS AND PROBLEMS 

1. Distinguish between soil organic matter, humus, and soil biomass. 

2. Give representative SOM contents for Mollisols, Aridisols, Oxisols, and His- 
tosols. Justify the SOM content of each in terms of the factors of soil formation. 

3. Explain why the SOM content of soils in a given climatic zone tends to be higher 
in fine-textured soils than in coarse-textured soils. 

4. Describe the overall decay process of SOM, including discussions of the or- 
ganisms involved, the decomposition products, and the time necessary for the 
process. 

5. If increased SOM contents are so beneficial to soils, why don't farmers manage 
their soils to increase SOM? 

6. What are the chemical properties of humus that make it special? 

7. How does SOM contribute to the chemical, physical, and biological properties 
of soil as a medium for plant growth? 

8. How may SOM alter micronutrient and trace metal availability in soils? 

9. What is the buffering capacity of humus expressed in terms of CaC03 equiva- 
lent? Assume 50% dissociation of humus acidity over the pH range of 4 to 7. 

10. How may SOM affect pesticide recommendations, and why? 

11. Explain how only a few percent organic matter can exert a profound influence 
on soil properties. 

12. Assuming that layer silicates and organic matter exist independently in a soil, 
calculate a reasonable cation exchange capacity of a soil containing 40% mont- 
morillonite and 3% organic matter. Repeat the calculation for a soil that 

(a) Contains 40% kaolinite and 3% organic matter. 

(b) Contains 20% kaolinite and 1.5% organic matter. 
How realistic are such calculations? 

13. Is SOM amphoteric? Explain. 

14. What are the primary functional groups responsible for charge development in 
SOM? 



7 



WEATHERING AND SOIL 
DEVELOPMENT 



Rocks formed beneath the earth's surface are unstable when raised to the surface 
where they contact water, O2. CO2, and organic compounds. Soil is formed when 
the ions in rock minerals at the earth's surface change, or weather, to more stable 
chemical stales. The change to increasing stability is slow and goes through many 
steps. Soil development, in the chemical sense, is roughly synonymous with weath- 
ering. This chapter discusses the general weathering reactions due to the effects of 
water, O2, and CO2 that create soil solids and the soil solution. 

The particle surfaces and soil solutions created by weathering tend to be more 
similar chemically than the composition of the parent minerals. This relative unifor- 
mity is of great benefit to the development and maintenance of life. The small vari- 
ation among different soils and soil solutions, however, affects plant growth and is 
often emphasized because of humanity's concern for maximum or optimum growth 
of agricultural plants. Compared to the vagaries of weather, pests, prices, and other 
production factors, soil chemical variability is relatively small and within a farmer's 
control. 

Weathering of igneous and metamorphic rocks changes these dense solids into 
unconsolidated particles whose surfaces and newly formed particles often differ 
markedly from the chemical composition and structure of the parent minerals. The 
changes during weathering of sedimentary rocks is less striking. The boundary be- 
tween sedimentary rocks and soil is often physically and chemically diffuse. 

Table 7.1 shows the composition of common soil parent materials. The crystal 
structures and ion valences in rock minerals are stable at the conditions under which 
the rocks formed. The physical conditions of erosion, freezing and thawing, glacia- 
tion, healing and cooling, and root growth at the earth's surface break rocks apart, 
which exposes more surface for chemical weathering. A bigger change in the rock 
minerals, however, results from the new chemical conditions: exposure to water, oxy- 
gen, carbon dioxide, and organic compounds. For sedimentary rocks, weathering is 

172 



WEATHERING AND SOIL DEVELOPMENT 



173 



Table 7.1. Average compositions of several parent material rocks 





Grvmodiorite" 












(Granitic) 


Basalt" 


Shale 6 


Sandstone* 


Limestone 


Compound 


(%) 


(%) 


(%) 


(%) 


(%) 


Si0 2 


65.1 


49.3 


58.1 


78.3 


5.2 


K 2 6 


2.4 


1.2 


4.3 


1.4 


0.04 


Ti0 2 


0.5 


2.6 


0.6 


0.2 


0.06 


Al 2 3 


15.8 


14.1 


15.4 


4.8 


0.8 


Fe203 


1.6 


3.4 


4.0 


1.1 


0.5 


Feb 


2.7 


9.9 


2.4 


0.3 


— 


MgO 


2.2 


6.4 


2.4 


1.2 


7.9 


CaO 


4.7 


9.7 


3.1 


5.5 


42.6 


Na 2 


3.8 


2.9 


1.3 


0.4 


0.05 


H 2 b 


1.1 


— 


5.0 


1.6 


0.8 


p 2 o 5 


0.1 


0.5 


0.17 


0.08 


0.04 


SO3 


— 


— 


0.6 


0.07 


0.05 


CO z 


— 


— 


2.6 


5.0 


41.5 


Total 


100 


100 


100 


100 


100 



"From G. W. "I'yrcll. 1950. The Principles of Petrology. Dutlon, New York. 

6 From F. J. Petlijohn. 1957. Sedimentary Racks, 2d ed. Harper & Row. New York. 



due to a change in those chemical conditions. The milieu in which sedimentary racks 
form is much closer to that of soil conditions than to the conditions of igneous and 
mctamorphic rock formation. Nonetheless, the weathering of sedimentary rocks is 
also due to exposure to water, 2 , and C0 2 , but at concentrations different from 
those in which the rocks formed. 

The major reaction that weathers minerals is the strong tendency of ions in solids 
to dissolve in water (Chapter 3). In addition to the energy of hydration released by 
dissolving in water, an ion's entropy increases as it is freed from a rigid structure into 
the aqueous solutions. Entropy increases further with increasing dilution in the water. 
Because the Gibbs free energy decreases with increasing entropy, because dilution 
also increases entropy, and because substances strive to reach their lowest energy 
state, an ion's most stable state is at infinite dilution in water or a solid. The amount 
of water in soils is limited to the thin film on soil particle surfaces, so that the entropy 
increase due to dilution is limited to that extent. 

After dissolving into the soil solution, some ions combine to create new solids 
that are stable under those soil solution conditions; other ions are leached away. This 
separation means that the composition of the soil solution changes during soil devel- 
opment and may change enough to cause the first-formed soil minerals to redissolve 
and then portions of them reprecipitate as still other minerals. Because minerals are 
stable over a range of composition, they change stepwise toward increased stability, 
while the soil solution's composition changes slowly and unidirectionally, but not 
continuously, during weathering. 



174 WEATHERING AND SOIL DEVELOPMENT 

Some minerals remain virtually unweathered despite their inherent instability, be- 
cause their dissolution rate in water is exceedingly slow. Quartz particles larger than 
several micrometers in size (fine silt) remain in soils for so long that quartz appears to 
be the most stable state for soil silicon. When finely divided into clay-sized particles, 
however, quartz persists only slightly longer in soils than does clay-sized feldspar. 
Feldspar disappears from the sand and silt fractions relatively rapidly. 

Slow reactions only delay the inevitable time when the unstable minerals either 
dissolve or form a new solid phase. New solids in soils sometimes form by recrys- 
tallization of another mineral, entirely in the solid phase. Much more frequently, 
new solids form by dissolution of the old mineral and subsequent precipitation of 
part of the solute ions, often on the surface of the old mineral. When part of the so- 
lutes reprecipitate, the overall process is called incongruent dissolution. Congruent 
dissolution is complete dissolution without subsequent reprecipitation of part of the 
original, for example, NaCl and CaCC>3 dissolving in water. Congruent dissolution 
is rare in nature. Soil weathering reactions, with the exception of limestone parent 
material, are incongruent dissolution because part of silicates remains behind as clay 
minerals and amorphous solids. 

Whether an ion remains in solution or precipitates depends on the sum of the 
heats of formation (enthalpy AH) of chemical bonds and the energy changes (en- 
tropy AS) associated with the randomness of ion movement and position. The AH 
values include the changes in energy between bonds in the old and new solids and the 
energies of ion-waler bonds. The change in randomness represents the freedom of 
ions to diffuse through water rather than being constrained within a crystal. Although 
an ion's water ligands have slightly less randomness than other water molecules, the 
net effect of weathering is increased randomness and hence increased AS. 

The AS change during ion dissolution tends to be similar for each ion and is rel- 
atively small. The AH of formation of new secondary minerals, in contrast, varies 
widely. The resulting driving force AC, therefore, differs widely between different 
ions. Ions that form weak chemical bonds with other ions (slightly negative AH, 
usually monovalent) tend to remain in aqueous solution, while strongly bonding ions 
(highly negative AH, usually polyvalent) tend to reprecipitate. Ions remaining in the 
soil solution are much more easily leached from soils and are therefore considered 
weatherable. The chemical states of reprecipitated ions also change during weather- 
ing. Because these ions remain in the soil, however, they are considered resistant to 
weathering. 

To illustrate, imagine a drop of pure water falling on the surface of albite 
(NaAJSijOg) feldspar. This igneous mineral is unstable in water at room temper- 
ature and pressure. The first ions or molecules to dissolve achieve a high degree of 
randomness because they can roam in the droplet unhindered by other ions: 

NaAlSi 3 Og + 4H 2 + 4H + = Na + + Al 3+ + 3Si(OH) 4 (7.1) 

albite soluble silica 

The next ions to dissolve find successively less randomness because other ions are 
already present in the water droplet and restrict their freedom of movement. Hence, 



WEATHERING AND SOIL DEVELOPMENT 175 

the AS change per dissolving ion progressively decreases. The AH of the albite 
bonds and of the bonds of the later minerals that precipitate are unchanged, so disso- 
lution slows and eventually stops. 

Aqueous solubility is governed by the counteracting effects of the AS of disso- 
lution versus the AW of chemical bonding in solids. The AH of Na + compounds 
tends to be small. Therefore, AS prevails and Na + remains in solution. The AH 
values of Al and Si bonds with O 2- and OH~, on the other hand, are large, so the 
aqueous solubility of those compounds is low, and the Al 3+ and Si(OH) 4 concentra- 
tions soon reach their solubility maxima in water. Soil solution conditions are often 
such that Al 3+ and Si(OH) 4 precipitate as kaolinite 

Al 3+ + Si(OH) 4 + 5H2O = 3H+ + i Al 2 Si 2 5 (OH)4 (7.2) 

kaolinite 

and Na + and some Si(OH) 4 remain in solution. At lower Si(OH) 4 concentrations, 
the AH of gibbsite predominates 

Al 3+ + 3H 2 = 3H+ + AI(OH) 3 (7.3) 

gibbsile 

so that Na + and Si(OH) 4 remain in solution. When the water film is displaced by 
a second droplet of water, the Na + and Si(OH) 4 are removed and weathering con- 
tinues. Leaching of kaolinitic soils eventually lowers the aqueous Si(OH) 4 concen- 
tration enough that kaolinite becomes unstable, degrades to gibbsite, and frees the 
remaining silicon: 

AI 2 Si20s(OH)4+5H 2 = 2Si(OH) 4 +2Al(0 H)j (7.4) 

kaolinite soluble gibbsite 

silica 

Continuing this simple view of weathering and soil development, the dissolution 
of feldspars and the formation of secondary clay minerals (Eqs. 7.1 and 7.2), tend 
to control soil pH in young soils. One mole of hydrogen ions is consumed per mole 
of albite weathered and the soil pH is slightly alkaline. When the feldspar content or 
rate of feldspar weathering decreases, Al 3+ and Fe 3+ hydrolysis reactions produce 
the acidity characteristic of moderately weathered soils (Eq. 7.3). In the final stages 
of soil development, reactions like Eq. 7.4 and the leaching of H + return soil pH to 
near neutrality. 

The albite example illustrates several general points about weathering: (1) Weath- 
ering of igneous and metamorphic minerals releases considerable alkali and alkaline 
earth cations during the initial transition from rock to soil. Some of these cations are 
retained by plant absorption and cation exchange, but most are lost immediately from 
the soil. (2) Weathering releases considerable silica to the soil solution. Much of the 
silica leaches from the soil. The remainder reacts to form the secondary minerals — 
kaolinites, smectites, chlorites — common in soils but transitory on a geologic time 
scale. (3) Aluminium (and Fe, Ti, and Mn) hydroxides are insoluble and tend to accu- 
mulate in soils. (4) Weathering initially produces some alkalinity. And (5) the second 
stage of weathering or soil development produces acidity. 



176 



WEATHERING AND SOIL DEVELOPMENT 



Slightly weathered 



Moderately weathered 



Strongly weathered 



Neutral to 
slightly alkaline 



Slightly alkaline 



CaC0 3 
accumulation 



7 



Soluble salt 
accumulation 



Parent material 



Organic 



Slightly acid 



Neutral to 
slightly alkaline 



Parent material 




Organic 



Neutral to 
slightly acid 



Highly acid 



Parent material 



FIGURE 7.1 . Schematic progression of basic and acidic zones through soils during soil devel- 
opment. This sequence also represents soil profiles from arid to humid to humid tropical regions. 

Figure 7.1 represents an idealized course of weathering in a soil profile. The ba- 
sicity and acidity are emphasized because pH is an easily measured indicator of the 
stale of weathering. The zones of basicity and acidity leach down through the soil 
profile during development. The alkali cations released during breakdown of the par- 
ent material accumulate in a narrow, rarely noticeable zone. This is followed closely 
by a much more obvious band of CaC03 accumulation. 

In slightly weathered soils, the surface soil pH is slightly alkaline. When the con- 
centration of exchangeable alkali and alkaline earth cations retained on clay surfaces 
and in soil solution decreases, the soil becomes acidic as the hydrolysis reactions of 
Al 3 "'" and Fe 3+ become evident. When the hydrolysis reactions near completion, the 
surface soil pH returns to near neutrality. Ultimately, acid production ends even in 
the subsoil. The entire profile returns to nearly neutral pH and the residual soil is rich 
in the Al and Fe hydroxyoxide products of weathering. 

The soil profiles in Fig. 7. 1 are also somewhat typical of soil profiles found in arid 
(slightly weathered), humid (moderately weathered), and humid tropical (strongly 
weathered) regions. The course of soil weathering is the same in all climates, but 
in humid regions weathering rates are sufficiently faster than the rates of erosion 
and physical mixing to allow strongly weathered soil profiles to develop. Soils in 
these regions are covered with vegetation, which slows erosion rates. Because of 
heat and rainfall, soils in the humid tropics pass quickly through the early stages of 
soil development. Soils in arid regions have slow rates of soil development because 
of the lack of water. Many arid region soils have relatively high wind and water 
erosion rates, because of sparse plant cover and because the rains come in violent, 
infrequent storms. 

The initial alkalinity from weathering is partially neutralized by the H2CO3 
formed in soil pores by root and microbial respiration: 



C0 2 + H 2 = H2CO3 = H + + HCOJ 



(7.5) 



WEATHERING AND SOIL DEVELOPMENT 1 77 

Neutralization of the bases released by weathering and the weak acidity of carbonic 
acid (pK ~ 5) favor continued weathering. The role of CO2 as the active agent of 
weathering has probably been exaggerated in the past. The abundance of water and 
the temperature more often control the rate of soil weathering than does the level of 
CCb. Carbon dioxide is always present in soil pores, whereas water is not. Weather- 
ing in desert and frozen soils is extremely slow. Equation 7.5 is driven to the right, 
that is, CCVs effect on the weathering rate increases, as the CO2 concentration rises 
to as much as several percent in soil pores during active root and microbial respira- 
tion, compared to 0.0033% v/v CO2 in the atmosphere. Desert and cold soils are low 
in CO2, because of less root and microbial activity, but still well above atmospheric 
concentrations. In practice, both water and high CO? levels are present during active 
soil weathering. 

Alkali and alkaline earth cations, halides, sulfate, and silica ions tend to remain 
in solution, rather than precipitate as secondary minerals. Despite the soil's ability 
to retain ions, these ions under normal soil drainage conditions eventually reach the 
sea. The K, Mg, and Si move more slowly than do Na, Ca, halides, and sulfate. 
Under arid to semiarid conditions, much Ca precipitates as lime CaC03 and some as 
gypsum CaS0 4 • 2H 2 0. 

The secondary minerals formed in soil from weathering products lend to be small 
in size and poorly crystallized to amorphous. They are primarily aluminosilicates 
and Al and Fe hydroxyoxides. These tiny crystals have large surface areas and tend 
to be charged because of unsatisfied chemical bonds within the crystals and at crystal 
edges. Large surface areas and unsatisfied bonds result in high surface free energies. 
Small particles therefore tend to dissolve and larger particles tend to grow at their ex- 
pense. This aging reaction is slow, however, because of the low solubility of the ions 
involved. In practice, soil-formed minerals rarely grow beyond colloidal (<2 /tin) 
size. Crystal growth is more evident for kaolinite. The weathering rates of smectites 
and chlorites, and leaching of their ions, are approximately as fast as their growth 
rates. Smectite and chlorite crystals largeT than 1 fun are rare in soils. 

Another important result of the unsatisfied bonds on the clay mineral surfaces 
is that they adsorb ions to balance the particle's charge. The unsatisfied charges are 
mostly negative, so mostly cations are adsorbed. Because they aTe held on the surface 
and not within the crystal, such ions can be exchanged for other ions. This slows the 
loss of ions from soils and retains the ions in states that are available for plant uptake, 
but ultimately the ions are lost. 

Weathering continues after the formation of secondary minerals because the sec- 
ondary minerals are stable only between certain concentration limits of soluble sil- 
ica, alkali and alkaline earth cations, and H + in the soil solution. As these solutes are 
leached away, the concentration changes make the initial secondary minerals (smec- 
tites, calcite, gypsum, etc.) unstable. As weathering progresses, these intermediate 
minerals weather further to still more stable chemical states. 

The effect of weathering reactions on the total composition of soils is illustrated 
by the data in Table 7.2. Three soils of increasing maturity, or degree of weathering, 
are compared to the average composition of igneous rocks. The comparisons are not 
exact because the parent materials of the soils differ from each other and from the 



178 



WEATHERING AND SOIL DEVELOPMENT 



Table 7.2. Composition of average igneous rocks and of three surface soils of 
increasing maturity 







Barnes 








Average of 


Loam 


Cecil Sandy 


Columbiana 




Igneous 


(South 


Clay Loam 


Clay (Costa 




Rocks 


Dakota) 


(North Carolina) 


Rica) 


Si0 2 


60 


77 


80 


26 


Al 2 o 3 


16 


13 


13 


49 


Fe 2 0-j 


7 


4 


5 


20 


TK> 2 ~ 


1 


0.6 


1 


3 


MnO 


0.1 


0.2 


0.3 


0.4 


CaO 


5 


2 


0.2 


0.3 


MgO 


4 


1 


<0.1 


0.7 


K 2 


3 


2 


0.6 


0.1 


Na 2 


4 


1 


0.2 


0.3 


P2O5 


0.3 


0.2 


0.2 


0.4 


SO3 


0.1 


0.1 


— 


0.3 


Total 


100.5% 


100.9% 


100.6% 


100.4% 



average igneous rock. The elemental contents are expressed as weight percent of 
the oxides, following an old geologic tradition, although only small fractions of the 
elements are actually present as simple oxides. Most are in more complex minerals. 
Expressing the elements as oxides allows the analyses to add up to 100%. More 
importantly, it emphasizes the importance of oxygen as the anion in rocks and soils. 
Even sulfates, phosphates, and carbonates are oxyanions in which the negative charge 
comes from oxygen ligands. The only other significant inorganic Iigand is sulfide, 
and sulfide minerals are unstable in aerobic soils. 

The first stage of weathering releases large amounts of Ca, Mg, Na, and K from 
the rock minerals, as illustrated in Table 7.2 by the change in composition from an 
igneous rock to the relatively young Barnes soil. With the exception of some Mg 
and K, these four cations are excluded from most pedogenic (soil-formed) minerals. 
Most of the alkali and alkaline earth cations remaining after the first weathering stage 
are in large unweathered mineral grains. Smaller fractions of Ca, Mg, Na, and K 
are retained by adsorption to negatively charged secondary mineral particles. These 
fractions are significant because they supply these essential macroelements to plants 
and soil microbes, they are subject 10 further leaching losses, and they control soil 
pH. 

Because the efflux of cations slows markedly after initial breakdown of rock min- 
erals, further stages are perhaps better called soil development than soil weathering. 
The difference in total composition between the immature Barnes soil and the rather 
mature Cecil soil is much less than that between the Barnes soil and igneous rock. 
The differences in soil maturity between the Barnes and the Cecil stages involve pri- 
marily the rearrangement of elements into secondary soil minerals. Rearrangements 
plus slight differences of total composition can create large differences in the avail- 



WEATHERING AND SOIL DEVELOPMENT 179 

ability of ions for plant growth. Phosphate is probably more available in the Barnes 
soil, for example, than in the Cecil soil, although the total amounts of phosphate are 
similar. Soluble plus exchangeable aluminium reaches phytotoxic concentrations in 
the Cecil soil. The slight differences in total elemental composition of the two soils 
can encompass a wide range of secondary mineral compositions. 

The phosphate content of soils tends to remain roughly constant during soil devel- 
opment. Phosphate is only slowly leached from soils, at about the same rate as silica 
loss, so the total phosphorus content of soils varies little with soil maturity. The form 
of phosphate, however, changes from predominantly apatite (Ca5(OH,F)(P0 4 )3) in 
igneous rocks to Al(III) and Fe(III) phosphates in moderately to strongly weathered 
soils. 

Soil sulfur, nitrogen, and carbon are associated with the soil's organic fraction 
and are relatively independent of weathering. Their soil contents are related to bio- 
logical activity and climate. Thus, while sulfates are potentially easily leached from 
soils, plant and microbial uptake of sulfate and its incorporation into organic com- 
pounds tend to maintain the sulfur content of soils. Sulfur is also added to soils as 
fallout from natural and polluted air. In industrial regions atmospheric sulfur from 
coal combustion provides a large supplement to the plant's supply of natural sulfur. 
Plants and soils absorb sulfurous gases directly from the atmosphere and receive sul- 
fates dissolved in rain. Nitrogen is also present as NO, NO2-N2O4, and NH3 in air 
and can be directly absorbed by plants and soils or from the soil solution after being 
washed out of the atmosphere. Fn addition, the reduction of atmospheric N2 to amino 
acids and proteins by special soil and plant bacteria is an important nitrogen input to 
all soils. 

Soil development involves a steady loss of silicon. Unfortunately, this is not read- 
ily apparent from the composition of the Barnes and Cecil soils in Table 7.2. The 
parent materials of these soils are apparently silica-rich. The loss of silicon is evi- 
dent, however, from the low SiC>2 content of the highly weathered Columbiana soil. 
As the SiC>2 content of this soil decreased to less than half the average for igneous 
rocks, the AI2O3 and Fe2C>3 contents increased threefold. The silica loss is from dis- 
solution, not erosion. The solubilities of Fe, Al, Ti, and Mn hydroxyoxides are much 
lower than the solubility of silica. These hydrous oxides, or sesquioxides, are more 
stable than the secondary silicates that might have formed earlier in this soil. As- 
suming that the Columbiana parent material is igneous rock, the threefold increase 
of hydroxyoxide content means that two-thirds of the original parent rock dissolved 
into the soil solution and has been lost, a loss of about 30 million kg ha - ' for each 
remaining meter depth of soil. 

The loss of solutes during the initial stages of weathering is even more obvious 
from the composition of the clay fraction than from the change of the total soil. 
Clay particles more accurately reflect the soil's chemistry; sand and silt particles are 
largely vestiges of the parent material. Table 7.3 compares the composition of the 
average igneous rock to that of the major silicate clay minerals that form in soils; 
the Al, Fe, Mn, and Ti hydroxyoxides are not included. The silicate clay minerals 
and the hydrous oxides contain much less Si, Ca, Mg, Na, and K than do igneous 
rocks. 



180 



WEATHERING AND SOIL DEVELOPMENT 



Table 7.3. The composition of soil clay minerals 9 compared to the average of igneous 
rocks 







Average of 


Hydrous 


Montmoril- 










Igneous 


Mica 


lonite 


Kaolinite 


Allophane 






Rocks 


(Scotland) 


(France) 


(Virginia) 


(Belgium) 


Compound 




(%) 


(%) 


(%) 


<%) 


(%) 


SiO> 




60 


49 


51 


45 


34 


Al 2 3 




16 


29 


20 


38 


31 


Fe 2 C>3 




7 


3 


0.8 


0.8 


Trace 


MgO 




4 


1.3 


3.2 


0.1 


— 


CaO 




5 


0.7 


1.6 


0.1 


2.3 


Na 2 




4 


0.1 


0.04 


0.7 


— 


K 2 




3 


7.5 


0.1 


0.1 


— 


H 2 (< 105° 


C) 


— 


3.2 


15 


0.6 


13 


H 2 (> 105° 


C) 


— 


6 


8 


14 


20 


Tola! 


99 


100 


100 


100 


100 



"l-'roin E. T. Dcgens. 1965. Geochemistry of the Sediments. Prcnticc-Hall, Englewood Cliffs, NJ. 



Aluminium, on the other hand, accumulates in the clay mineral fraction because 
it forms insoluble aluminosilicates and hydroxyoxides. The Al remains behind in the 
soil as other ions leach away. Iron also accumulates in soils but this is not appar- 
ent from Table 7.3 because the silicate clay minerals, with the exception of hydrous 
mica, are low in Fe. Iron precipitates in soils only as hydroxyoxides. Hydrous mica is 
altered parent material and is not reconstituted from the soil solution as are kaolinite, 
montmorillonite, and allophane. The <105° C water in Table 7.3 is, roughly speak- 
ing, adsorbed water; the >105° C water is hydroxy 1 ions and water within crystal 
structures. 

Small amounts of weathered solutes reprecipitate in lower soil horizons. Exam- 
ples include clay accumulation in the B horizon, "silica pans" (impermeable layers 
of soil particles indurated with silica), and the wide-spread "caliche" horizons of 
CaC03 accumulation in arid regions. Most solutes, however, reach the sea, where 
precipitation of other secondary minerals removes most of the weathered solutes 
except Na + , Cl~, and Mg 2+ . Marine sediments, in turn, are slowly converted into 
igneous, metamorphic, or sedimentary rocks, which form new soil parent material. 
Such element recycling has circulated ions many times from land to sea during the 
earth's history. 

Although the net effect of weathering is the loss of soluble components from the 
soil, the course of weathering is by no means continuous or unidirectional. The soil 
solution can flow upward during dry seasons. Plant uptake and decay deposit ions on 
the soil surface. The rates of ion absorption by plants during nutrient cycling are far 
greater than the rates of weathering. Because plant uptake greatly reduces weathering 
losses, large-scale losses of ions from soils usually occur only during periods of high 
rainfall or limited plant growth due to overgrazing, forest clearing, or fires. Chemical 
elements are also moved physically in the soil profile by "vertical mulching"; by soil 



STABILITY OF PARENT MATERIAL MINERALS 181 

fauna, such as worms, termites, and ants; and by uprooting of fallen trees. Vertical 
mulching occurs in monlmorillonitic soils (Vertisols) that can form extensive vertical 
cracks as deep as 1 m in the dry season. Soil particles break from the walls of the 
cracks and fall to the bottom. When wetted, the fallen particles swell and force the 
overlying soil upward. The result is a slow, continual mixing of the surface meter of 
soil. 

Another example of weathering reversal is salt accumulation due to impeded soil 
drainage or seawater inundation. Weathering under these conditions reverses in the 
sense that the secondary minerals formed are chemically similar to igneous and sed- 
imentary rock minerals and are unstable under well-drained, oxidative conditions. 
The chemistry of soil development deals with the degradation of parent minerals and 
the formation of secondary minerals over a wide range of chemical conditions. 



7.1 STABILITY OF PARENT MATERIAL MINERALS 

The major minerals of igneous rocks are, in decreasing order of general abundance, 
feldspars (MAlSi 3 8 ), quartz (Si0 2 ), and biotite (K(Mg,Fe) 3 AlSi 3 Oi (OFi)2) and 
muscovite (KAhSbOioCOHh) micas. The feldspars include orthoclase and micro- 
cline (both KAlSi 3 Og) and the plagioclase series ranging from albite (NaAlSi 3 Og) to 
anoilhite (CaAl2Si20g). Other minerals in igneous and sedimentary rocks are gen- 
erally present in lesser amounts and have chemical compositions similar to those 
above. Granitic or acid (>66% SiOi) igneous rocks are richer in silicon and potas- 
sium, and poorer in magnesium and iron, than basaltic or basic igneous (45 to 52% 
SiCb) rocks (Table 7. 1 ). Olivine ((Mg,Fe)SiO_ 3 ) is a distinctive component of basaltic 
rocks. 

Metamorphic rocks contain the above minerals and chemically similar variants. 
Metamorphic rocks are occasionally rich in Mg minerals, such as the pyroxenes 
(approximate composition Ca(Mg, Fe)Si206) and the chemically similar but more 
complex augites. Serpentine metamorphic rocks are even richer in Mg because of 
antigorite (Mg 3 Si 2 Os(OH) 4 ). 

Sedimentary rock materials have already passed through some weathering before 
the rock is formed. Their composition represents depletion of weatherable elements, 
as in sandstone, or selective accumulation, as in shale and limestone (Table 7.1). 
Sandstones are mostly mechanical accumulation of quartz grains and are depleted of 
virtually all other elements. Shales, which are fine-grained sedimentary rocks, tend 
to form in regions of some chemical and particulate accumulation. They are richer 
than sandstones in K and Mg aluminosilicates, Fe, and Ca and Mg carbonates. The 
Na content of sedimentary rocks is generally much lower than that of igneous rocks, 
although some marine sediments contain entrapped Na. 

The mineral structures of igneous minerals are varying organizations of silicon- 
oxygen tetrahedra and aluminium-oxygen tetrahedra and octahedra. The stability of 
igneous minerals also applies to the weathering of chemically and structurally similar 
minerals in metamorphic and sedimentaiy rocks. 



182 



WEATHERING AND SOIL DEVELOPMENT 



Olivine 
\ 
Hypersthene 



Increasing 
stability to 
weathering 



Augite 
\ 
Hornblende 

X 

Biotite mica 



Ca plagioclase 

/ 

Ca-Na plagioclase 

Na-Ca plagioclase 

Na plagioclase 

/ 
K feldspar 

T 
Muscovite mica 

T 

Quartz 



Order of 
crystallization 
from magma 



Decreasing 
energy of 
formation 



FIGURE 7.2. Stability to weathering of some minerals in igneous and metamorphic rocks. 
(Adapted from S. S. Goldich. 1938. J. Geol. 46:38.) 



The resistance of igneous minerals to weathering is the same as the order of crys- 
tallization from cooling magmas (Fig. 7.2). Minerals that are most stable at high tem- 
peratures, and that therefore crystallize first from the molten magma or lava, are the 
least stable at low temperatures. Such minerals, including calcic feldspars, olivine, 
and hypersthene ((Mg,Fe)Si03) tend to be rich in the water-soluble alkali and alka- 
line earth and Fe(II) ions. The weathering rate generally increases with increasing 
content of alkali and alkaline earth cations. 

A second chemical factor affecting mineral weatherability is the position of ions 
in the structure. The tetrahedra of Ca feldspars contain half Al 3+ and half Si 4+ . At 
room temperature, Al 3+ is more stable in octahedral coordination. The charge deficit 
created by the Al 3+ substitution is made up by Ca 2+ ions between the tetrahedra. The 
structural strain, the charge deficit in the tetrahedra, and concentrated Ca 2+ counter 
charge weaken the anorthite feldspar structure with respect to weathering relative 
to Na and K feldspars. In Na and K feldspars, only one-quarter of the tetrahedral 
positions are occupied by Al and that charge deficit can be locally neutralized by 
Na + or K + . Calcium feldspars are, therefore, the least stable feldspars under soil 
conditions. Potassium feldspars are more stable than Na feldspars, because K fits 
better between adjacent tetrahedra. 

A third chemical factor affecting mineral weatherability is the degree to which the 
tetrahedra are linked together. Increased linkage between tetrahedra means increased 
stability against weathering. Feldspars and quartz are three-dimensional networks of 
tetrahedra in which each of the four tetrahedral oxygens is a corner of another tetra- 
hedron. This maximum sharing of oxygens produces considerable stability. Hence, 
quartz is very persistent in soils, if it is sand- or silt-sized. Feldspars would also be 
resistant to weathering if not for their Ca, Na, and K contents and the other fac- 
tors described above. These are unfavorable for feldspar stability and counteract the 
stabilizing effect of tetrahedral linkage. 

The sharing of tetrahedral oxygens in other silicates varies from the maximum 
sharing in quartz and feldspar to the complete independence of the tetrahedra in 
olivine. Olivine weathers very rapidly when exposed to water and air. Pyroxenes 



IONIC POTENTIAL 183 

share two corners of each tetrahedron to form parallel single chains of tetrahedra. 
Amphiboles share three corners to form parallel double chains. Pyroxenes and am- 
phiboles are appreciably more resistant to weathering than olivine. In micas the 
sharing is more complex and similar to the structure of smectites. Two sheets of 
silicon-aluminium tetrahedra are joined together by an included layer of aluminium 
or magnesium-iron octahedra. Charge imbalance within these three-layer sheets is 
balanced by K + (and occasionally NH^ ) on the outer surface of the silica sheets. 
High-aluminium (dioctahedral) muscovite mica is more stable than high-iron and 
high-magnesium (trioctahedral) biotite mica, in part because of a better fit of the in- 
ternal aluminium octahedral sheet. Chapter 5 gives details of mineral composition 
and structure. 

A fourth factor of mineral stability is the Fe(II) and Mn(II) contents. In aerobic 
soils the presence of these ions increases weathering rates because their oxidation 
to Fe(III) and Mn(III,IV) creates charge imbalances. In anaerobic soils, on the other 
hand, minerals containing the oxidized ions are unstable. 

Figure 7.2 summarizes the relative weathering rates of major minerals in igneous 
and metamorphic rocks. Actual weathering rates depend also on soil temperature 
and moisture, particle size, and planes of physical weakness (cleavage) in the crys- 
tal. The effect of moisture includes both the flow rate of soil solution past mineral 
surfaces and the composition of the solution. Solids dissolve more slowly if the so- 
lution already contains their constituent ions. High electrolyte concentrations, on the 
other hand, can maintain higher ion concentrations at equilibrium because of lower 
activity coefficients and because of complex-ion and ion-pair formation. 

Smaller particles weather more rapidly, but the size effect is great only when the 
particles are less than several micrometers in size. Cleavage planes allow particles to 
be more easily broken apart. Feldspars and micas, for example, have clearly defined 
cleavage planes. Particularly in the case of feldspars, the cleavage planes hasten the 
rate of mineral breakdown. 



7.2 IONIC POTENTIAL 

The geochemist Goldschmidt tried to predict the fate of ions released during the 
weathering of igneous minerals. He found that the ionic potential, the ratio of crystal 
ion radius to ion charge, indicated fairly well whether ions eventually would leach 
to the sea or remain behind in soils and sediments. The crystal (dehydrated) ion size 
is useful for predicting solution behavior, because the size indicates how strongly it 
can react with water, OH - , and O 2- . 

Figure 7.3 plots the ratio of crystal radius versus charge for selected ions. 
Oxyanions — sulfate, selenate, phosphate, arsenate, borate, molybdate, carbonate, 
and silicate — are represented by their central cations: S 6+ , Se 6+ , P 5+ , As 5+ , B 3+ , 
Mo 4+ , C 4+ , and Si 4+ . The ions fall into three behavioral groups. Ions of high ionic 
potential, the alkali and alkaline earth cations and the halide anions, large univalent 
and divalent ions, are highly water soluble, easily weatherable, and leach readily 
from soils to the sea over geologic time. 



1 84 WEATHERING AND SOIL DEVELOPMENT 




♦3 
ION CHARGE 

FIGURE 7.3. Ionic potentials of important ions. 



Ions oi" intermediate ionic potential, such as aluminium and the transition metals, 
include most of the elements in the periodic table, although only Al, Fe, Ti, and Mn 
are present in appreciable amounts in soils. Ions in this group are of intermediate size 
and charge. They tend to polarize their associated water molecules and to repel H + 
sufficiently to precipitate as insoluble hydroxyoxides. Such ions are released during 
weathering of silicates but, because of the low solubilities of their hydroxyoxides, 
are leached only very slowly from soils. Ions of higher and lower ionic potentials 
migrate faster, so this middle group of ions tends to remain behind and accumulate 
in soils. Hence highly weathered soils, such as Oxisols or Ferrasols, contain high 
amounts of Al, Fe, Ti, and Mn oxides. 

The ions with the lowest ionic potentials are oxyanions formed by the smallest, 
most highly charged cations — SQ^ - , NOJ , CO3 - , and so on. These cations repel 
several protons from associated water molecules to form permanent oxide ligands. 
Such oxyanions are water soluble by geochemical standards and hence tend to be 
leached from soils. Phosphate and silicate are the least soluble members of this 
group. They lie near the boundary between soluble oxyanions and insoluble hydrox- 
ides. The loss of borates and silicates, but not so much of phosphate, is characteristic 
of moderate to advanced stages of soil development. 

The ionic potential can be used to predict general soil chemical behavior of ions, if 
combined with a knowledge of the ion's oxidation-reduction behavior and of the soil 
chemistry of chemically related ions. The ionic potential describes best the overall 
process of rock weathering to form sediments, the process for which it was intended. 
It is somewhat less successful in explaining the soil chemical behavior of some ions. 
Magnesium is generally weatherable in soils. Sodium is retained much more weakly 
than is K" 1 . The strong preference of particular sites on layer silicate minerals for K + 
cannot be accounted for by ionic potential. Similarly, Ca 2+ is less strongly retained 



RATES OF WEATHERING AND SOIL DEVELOPMENT 1 85 

than Mg 2+ . The difference is due to the incorporation of Mg 2+ into secondary miner- 
als. Calcium released by weathering remains in the soil solution as an exchangeable 
cation. This difference is not accounted for by the ionic potential. Comparing the 
exchangeable forms only, Mg 2+ is usually adsorbed less strongly than Ca 2+ . The 
exceptional preference of vermiculite for Mg 2+ is discussed in Chapter 8. 



7.3 RATES OF WEATHERING AND SOIL DEVELOPMENT 

Temperature and moisture are the major environmental variables affecting weather- 
ing rates. Assuming similar chronological ages and parent materials, the difference 
in composition of the South Dakota, North Carolina, and Costa Rican soils of Ta- 
ble 7.2 illustrates the effects primarily of temperature. Weathering is much faster in 
the warm climate of Costa Rica than in the cold winters and short summers of South 
Dakota. North Carolina's climate is intermediate between the two. 

The rate of water movement through soils determines the rate at which weathered 
solutes are removed from the vicinity of soil particles. Weathering can continue even 
when the rate of movement is slow, such as in poorly drained soils. Lack of water, 
however, almost totally arrests soil development. Desert soils can be old chronologi- 
cally, yet young in the sense of soil development. 

Jenny (1941) proposed that soil development be regarded as the result of five 
soil-forming factors: climate, topography, biosphere, parent material, and time. In a 
qualitative sense, the weathering rate is related to these factors by 

Aweathering . ,. , . , , . 

— J (climate, topography, parent material, biosphere) (7.6) 

Atvme 

Converting this expression into a quantitative equation, however, is beyond our 
present capabilities. None of the four factors has been adequately described numer- 
ically. Climate, for example, is an ill-defined integration of the intensity, duration, 
and seasonal distribution of temperature, moisture, and evaporation. Deposition 
of airborne salts and dust and parts of erosion should also be included as subtle 
parts of climate. The parent material factor includes chemical composition, mineral 
composition, crystal size, and rock fabric (structure). 

The relative importance of each factor in Eq. 7.6 also varies with local and re- 
gional conditions. In a peat bog or on a steep mountain slope topography clearly 
has a prominent role. For a young alluvial soil, on the other hand, parent material is 
usually the dominant factor. In desert and polar soils, the biosphere's role is compar- 
atively small. 

The soil-forming factors also are interdependent. The biosphere clearly depends 
on climate. Topography may control drainage, but soil water movement is also af- 
fected by soil texture, which is derived in part from the crystal size of the parent 
material. Topography includes aspect (the direction of slope), which contributes to 
the microclimate of the soil. A north- facing slope (in the northern hemisphere) is 
cooler and moister than a south-facing slope. This affects the nature and distribu- 
tion of plants, as well as the soil directly. The biosphere greatly affects soil-water 



186 



WEATHERING AND SOIL DEVELOPMENT 



relations, produces organic molecules that react with soil particles, dominates car- 
bon and nitrogen chemistry and other oxidation-reduction processes of soils, and 
increases the CO2 concentration of the soil air and soil solution. These interactions 
greatly increase the complexity of Eq. 7.6. 

The composition of water draining from soils is an index of the rates of mineral 
weathering and ion leaching. The losses of several elements from the earth's conti- 
nents are shown in Table 7.4. The values are the products of river water composition 
and river flow rates. Although they include some amounts dissolved from deeper 
sediments, the amounts are global indices of soil weathering and losses. The alkali 
and alkaline earth cations are lost mainly as dissolved ions. Silicon, aluminium, and 
iron are lost mainly by erosion as suspended sediments rather than by chemical dis- 
solution, although silica dissolution is appreciable. Other estimates of the Ca, Mg, 
K, N, and S losses by leaching are also in the range of 2 to 20 kg ha -1 yr -1 (50 to 
1000 mol ha -1 yr" 1 ). 

The rates in Table 7.4 and the soil composition data in Table 7.2 can be combined 
to calculate the residence time, or turnover rate, of ions in soils. Residence time is 
the total soil content divided by the loss rate and is the inverse of the weathering 
rate. The soil's sodium content is approximately 0.4%, or 50 000 kg ha _i m _1 . The 
sodium loss of 300 mol ha -1 m _l , or 7 kg ha -1 m~', indicates that sodium's resi- 
dence time is about 7000 years in the surface meter of soils. This estimate disregards 
atmospheric sodium input to the soil and also disregards the contribution of sediment 
weathering to the sodium losses in Table 7.4. If sediment weathering contributes half 
the sodium content of the world's rivers, sodium's residence time in soils is on the 
average about 1 5 000 years. 

Similar calculations disclose that calcium's residence time in soils is about 2500 to 
5000 years. The Na and Ca weathering rates seem exceedingly fast but are, of course, 



Table 7.4. Weathering losses from the continents to the sea 3 (continental land 
area = 1.2 x 10 8 km 2 ) 











Soluble 






Sediment 


Element 


x 10 


12 moles 


yr 


-1 


moles ha ' 


y^ 1 


moles ha -1 yr -1 


Sodium 




3.6 






300 




130 


Potassium 




1.8 






150 




120 


Magnesium 




5.5 






460 




HO 


Calcium 




13 






1100 




130 


Silicon 




9 






750 




3300 


Aluminum 




— 






— 




930 


Iron 




— 






— 




300 


Sulfur 




1.9 






160 




— 


Chlorine 




2.7 






220 




— 


Phosphorus 




— 






2 




— 



"Mostly from Garrels, Mackenzie, and Hunt. 1975. Chemical Cycles and the Global Environment. W. 
Kauffman, Los Altos, CA. 



RATES OF WEATHERING AND SOIL DEVELOPMENT 1 87 

counterbalanced by similar rates of input. Soils have weathered at about these rates 
for 4.5 x 10 9 years and are still far from exhausted. 

The residence time can also give some idea of the rate of element cycling between 
soils and plants. Table l . I gives the soil's Na supply, relative to rate of plant uptake, 
as equivalent to about 5000 years. Sodium ions, therefore, cycle two to three times, 
on the average, through plants and soil before being leached from the soil. The 260- 
year plant supply of Ca in soil (Table 1.1) indicates that Ca cycles 1 to 20 times 
through plants before being leached from soils. 

The leached ions are replenished by atmospheric fallout of sea spray entrapped 
in rain and by the formation of new igneous rock at the ocean floor. Assuming that 
the Na and Ca concentrations in the ocean are constant, their ocean residence time 
equals the replenishment rate of soils. Ocean residence time, in turn, is equal to the 
concentration divided by the rate of input from the world's rivers. The Na residence 
time in the oceans is about 100 million years; the Ca residence time is about 1 million 
years; K and Mg are intermediate. Comparing the residence times of Ca in soils and 
oceans shows that 200 to 400 m of soil are weathered of their Ca content during the 
1 million years residence time of Ca in the oceans. 

The ions are replenished mainly by formation of new igneous rock, at a rate re- 
cently estimated at 12 x 10 9 m 3 yr _l or 30 Tg yr _l . The turnover rate of the entire 
crust of the earth is therefore about 200 million years; the entire earth's crust has been 
weathered and remelted into rock on the average about twenty times over the lifetime 
of the earth. The earth and its soil appear stable to us only because our residence time 
on the earth is so short. 



7.3.1 Acidity 

Increasing acidity of the soil solution hastens weathering. Discussions of weathering 
usually mention the H2CO3 of rainwater as both a significant input of acidity to soils 
and a significant factor in soil mineral weathering. The solubility of CO2 in water in 
equilibrium with the atmosphere is about 10~ 5 M. An annual rainfall of 1000 mm 
means an input of dissolved CO2 of 0.01 mole m~ 2 of soil surface. The amount of 
CO2 produced by biomass is much greater. The annual net carbon fixation by photo- 
synthesis is about 1 kg C m~ 2 of productive soil, or 80 mol C m -2 . Assuming that 
the amount of carbon respired by roots and soil microbes equals net photosynthesis, 
and that is a very conservative assumption, roots add 8000 times more CO2 to the 
soil than does rainwater. Decay of surface plant residues adds additional CCb, but 
most of this is liberated directly to the atmosphere. 

Other atmospheric components, such as HNO3 and NH3, are usually too dilute in 
nature to affect the acidity of rain except for the first droplets, which flush out the air. 
The SO2 and NO A (NO+NO2-N2O4) released by coal combustion, automobiles, and 
ore smelting, however, can significantly increase atmospheric acidity. These gases 
oxidize to H2SO4 and HNO3 in the atmosphere, on leaves, and at the soil surface. 
Acid rain of pH 4.5 or lower is common in industrial regions and may seriously affect 
plants, lakes, buildings, and perhaps the weathering rates of soil minerals. Acid rain 
has received less attention in recent years. The changes in lake acidity that caused 



1 88 WEATHERING AND SOIL DEVELOPMENT 

part of the concern were due in part to changing the pH measurement technique 
from indicator dyes to the glass electrode. Lake and stream waters are poorly pH 
buffered so the slight alkalinity of the indicator dye gave a higher value than the later 
measurements by the glass electrode. In addition, the abandonment of agriculture in 
northeastern North America meant the abandonment of liming. The soils returned to 
their native acidity and so did the water draining from these soils. Acid rain is a real 
phenomenon, but air pollution controls on power plants and vehicles are reducing its 
intensity. 

Although acid rain may increase the weathering rates of soil minerals, its ef- 
fect could be rather small. Plants and microorganisms absorb sulfate and nitrate 
anions and simultaneously excrete an equivalent quantity of OH - to maintain in- 
ternal charge balance. Plant absorption of these anions could thus tend to counteract 
the effect of acid rain on soil weathering. Benefits from such anion absorption by 
plants would accrue primarily in acid soils, where natural sulfate concentrations are 
relatively low. In neutral and alkaline soils, the high pH and high acid-buffering ca- 
pacities would counteract the effects of acid rain. In neither case would acid rain 
increase soil weathering rates to the degree once feared. 

7.3.2 Mechanisms of Mineral Decomposition 

The decomposition of several common soil minerals has been examined in the labo- 
ratory and probably represents the mechanism of mineral weathering in soils as well. 
The decomposition rale of montmorillonite is proportional to the H + concentration 
of the attacking solution: 

— A (mont) , 

\ - W + ) (7.7) 

At 

where — A(monf)/ At is the rate of disappearance of montmorillonite with time. This 
type of rate equation is thought to result from the free expansion of montmorillonite 
mineral sheets, so that the mineral's total surface is susceptible to attack. Hence, 
the decomposition is independent of the amount of montmorillonite, a pseudo-zero- 
order reaction. The rate depends only on the H + concentration as long as the H + 
concentration is in the range of soil solution acidities, that is, if the H + concentration 
is low relative to the montmorillonite concentration. 

The rate equation for kaolinite decomposition has the form 

-A tool) 

= A(kaol)(H + ) (7.8) 



At 

Hence, the rate of kaolinite decomposition depends on both the acidity and the kaoli- 
nite concentration. As kaolinite decomposes, its rate of decomposition decreases, and 
complete disappearance should theoretically require infinite time. Indeed, kaolinite 
is quite resistant to weathering. Kaolinite is nonexpanding so its exposed surface 
is small. Inasmuch as few soil minerals expand, Eq. 7.8 probably characterizes soil 
mineral weathering better than does Eq. 7.7. Smaller mineral particles tend to de- 
compose first, leaving behind the larger particles. The weathering rale thus dimin- 



RATES OF WEATHERING AND SOIL DEVELOPMENT 1 89 

ishes with time, so that small amounts of weatherable materials may persist even in 
highly weathered soils. 

As the remaining particles weather to progressively smaller sizes and large surface 
areas, the last remnants might be expected to decompose quite rapidly. This would 
be true if particle surfaces remained fresh and unweathered. In practice, however, 
the particle surfaces become coated with residues from the weathering process and 
with reprecipitated secondary minerals. The residue coating hinders decomposition 
of weatherable soil minerals, since weathering products must diffuse through this 
layer before they can dissolve in the soil solution. Weathering agents must diffuse 
in the opposite direction to attack the mineral. The weathered surface is a protective 
coating. If the soil weathering rate increases, the rate of release of alkali and alkaline 
earth cations and silica should increase. This, in turn, would leave a thicker layer 
of Al(OH)3 and FeOOH remaining on the mineral surface. The thicker layer is a 
negative feedback mechanism that reduces ion diffusion rates and the weathering 
rate. 



7.3.3 Time Sequence of Mineral Occurrence 

Weathering involves the movement of water through the soil profile and the gradual 
removal of mainly silica and alkali and alkaline earth cations. The flow of water 
usually prevents accumulation of soluble salts but is slow enough to permit Si, Al, 
and some Mg to reprecipitate as secondary minerals. Under such conditions, Jackson 
and Sherman (1953) proposed that the change with time of the material composition 
of soil clays is similar in many soils (Table 7.5). The clay fraction better reflects the 
time, chemical composition, and environmental conditions that existed during soil 



Table 7.5. Sequence of clay mineral distribution with increasing soil development 3 

Relative Degree of 

Soil Development Prominent Minerals in Soil Clay Fraction 

1 Gypsum, sulfides, and soluble salts 

2 Calcite, dolomite, and apatite 

3 Olivine, amphiboles, and pyroxenes 

4 Micas and chlorite 

5 Feldspars 

6 Quartz 

7 Muscovite 

8 Vermiculitc and hydrous micas 

9 Montmorillonites 

10 Kaolinite and halloysite 

1 1 Gibbsite and allophane 

12 Goethite, limonite, and hematite 

13 ' Titanium oxides, zircon, and corundum 

"Adapted from M. L. Jackson and G. D. Sherman. 1953. Advances in Agronomy, 5:221-319. 



1 90 WEATHERING AND SOIL DEVELOPMENT 

formation than does the soil as a whole. The sand and silt fractions are usually relics 
of the soil parent material. 

The mineral groups of Table 7.5, when present in the clay fraction, indicate pro- 
gressively increasing stages of soil maturity. A given suite of minerals may not nec- 
essarily dominate the clay fraction, but its presence in detectable amounts is a fairly 
reliable indicator of the degree of soil weathering and development. The criterion for 
a mineral's presence normally is whether the mineral is detectable by x-ray diffrac- 
tion. Small amounts or poor crystallinity can lead to detection problems. Jackson and 
Sherman's clay mineral groups that are characteristic of increasing soil maturity are 
as follows: 

1 . Soluble salts— halite (NciCl) and gypsum (CaSC>4 ■ 2H2O), as well as sulfides 
(pyrite, FeS2) in soils reclaimed from seas or swamps. These minerals readily 
dissolve in percolating water or, in the case of the sulfides, are readily attacked 
by oxygen. Saline and sodic ("alkali") soils are examples of this category. 

2. Calcite (CaC0 3 ), dolomite (CaMg(COs) 2 ), and apatite (Ca 5 (F,OH)(P0 4 ) 3 ). 
Carbonates of clay size are rather rapidly leached from humid soils. In arid 
regions, calcite accumulates. The phosphate remains in the clay phase after 
apatite decomposition, although in the form of Ca phosphates in alkaline soils 
or Al and Fe phosphates in acid soils. 

3. Olivine ({M%,Fe)iSi04) and the feldspathoids (amphiboles, mainly horn- 
blende, and pyroxenes). In these minerals Fe(II) can be oxidized, increasing 
the weathering rate. 

4. Primary layer silicates, biotite (K(Mg,Fe,Mn) 3 Si3AlOio(OH)2) from igneous 
and metamorphic parent materials, glauconite (K(Fe,Mg,Al)2Si 3 AlO]o (OH)2) 
from marine sediments, and magnesian chlorite (Mg,Fe)a(Si,Al)40jo(OH)8. 
Aluminium is a common substitute (up to 1 in 4) for silicon in the tetrahedral 
sheets. This is an unstable configuration for Al at room temperature, but is 
more than counterbalanced by the stability imparted by the sharing of oxygens 
between the tetrahedral and octahedral interlayers. Iron(II) in these minerals 
increases their instability with respect to oxidation. 

5. Feldspars. Albite (NaAlSisOg) and anorthite (CaAl2Si20a) are the end mem- 
bers of the plagioclase feldspar continuum, covering the whole range of Na,Ca 
mixtures. The greater the calcium content of plagioclase, the faster its rate of 
weathering. Orthoclase and microcline feldspars (both KAIS13O8) in the clay 
fraction weather at roughly the same rate as albite. 

These five groups are inherited from the soil's parent material and are considered eas- 
ily weatherable minerals when clay-sized. Inherited minerals that are more resistant 
to weathering in the clay fraction include: 

6. Quartz (S/02J. Quartz in the clay fraction is much more easily weathered than 
in the sand fraction, where it is the most abundant mineral. 



RATES OF WEATHERING AND SOIL DEVELOPMENT 191 

7. Muscovite (KAhfSijAOOiofOHh). Muscovite mica is considerably more sta- 
ble than biotite. Biotite contains oxidizable Fe(II) and some Mn(II), whereas 
muscovite does not. In addition, the dioctahedral (Al) layer of muscovite seems 
to fit much better between the two layers of silica-alumina tetrahedra, and 
hence is more stable, than the trioctahedral (Mg and Fe(Il)) layer of biotite. 

8. Interstratified or intermixed layer silicates, vermiculite (M + (Mg,Fe)3(St4- n , 
Al„)Oio(OH)2) and the hydrous (slightly weathered) micas. The M + repre- 
sents an exchangeable cation. Whether the minerals of this category are inher- 
ited from the parent material or are secondary products derived from inherited 
minerals is uncertain in many cases. 

The remaining categories of Table 7.5 contain secondary (soil-formed) minerals 
resulting from the weathering of primary minerals from soil parent materials. By the 
time these stages of weathering are reached, inherited minerals in the clay fraction 
have either disappeared or are present in only minor amounts. 

9. Montmorillonites or smectites (M + (Al,Mg)Si40io(OH)2)- These are the sec- 
ondary Mg- and Al-rich 2:1 layer silicates that can form in soils. The chemical 
composition can vary substantially, but the basic expansible structure remains 
essentially the same. 

10. Kaolinite and halloysite (Al2Si205(OH) 4 ). The 1:1 layer lattice kaolins are 
more stable with time than the 2:1 smectites. Kaolinite is a common compo- 
nent of soil clays and occurs in relatively high concentrations in moderately 
weathered soils. 

11. Aluminium hydroxyoxides. These include gibbsite (AKOHh), boehmite and 
diaspore (AlOOH), and allophane. The loss of silicon from soils leaves an Al- 
and Fe-rich residue in soil clays. 

12. Iron hydroxyoxides. These include goethite (FeOOH), limonite (a hydrated 
Fe(III) hydroxyoxide), and hematite (Fe203). 

13. Titanium oxides. These include anatase and rutile (Ti02), leucoxene (hy- 
drated, amorphous titanium oxide), and ilmenite (FeTi03), plus zircon (Z1O2) 
and corundum (AI2O3). Titanium and zirconium are so immobile in soils that 
members of mineral group are used as indicators of the amount of parent 
material that has weathered to produce a given volume of soil. 

The above sequence of mineral occurrence does not imply that all ions present 
in the soil progress from one weathering category to the next. Table 7.1 shows that 
only a small traction of the alkali and alkaline earth cations are retained by soils when 
igneous minerals first weather. Also, when montmorillonite weathers, only a fraction 
of the Si02 and exchangeable cations recombine with Al(OH)3 to form kaolinite. 
The remaining fractions are leached from the soil. Indeed, the first identifiable solid 
product of igneous mineral weathering is generally gibbsite. Whether the gibbsite, 
soluble silica, and cations remain apart or reprecipitate as smectites, kaolinite, and 
chlorite depends on the chemical composition of the soil solution and particularly on 



192 WEATHERING AND SOIL DEVELOPMENT 

the cation and Si(OH)4 concentrations. The sequence in Table 7.5 is primarily a time 
sequence, in the sense that cations and silica are increasingly lost from soils with 
time. The secondary minerals in each successive weathering stage of Table 7.5 are 
stable at lower cation and Si(OH) 4 concentrations, and lower soil pH, than those in 
the previous stage. 

Recent work has shown that Table 7.5 may be too rigid. Mineral occurrence over- 
laps considerably. Tables 7.6 and 7.7 are a more current view of mineral occurrence 
in soils. 



7.4 MINERAL FORMATION IN SOILS 

The formation of secondary minerals in soils generally results from the combination 
and addition of ions and molecules from the soil solution to the solid phase. This 
mechanism was originally given little consideration, because aluminium and silicon 
in solution did not appear to combine during laboratory experiments. Only relatively 
recently has the slow kinetics of such reactions been appreciated. Experiments that 
take slow reactivity into account and provide nucleation centers for crystal formation 
have shown that secondary minerals can precipitate from solutions containing the 
proper constituent ions and Si(OH) 4 . 

Formerly, soil minerals were thought to form by differential migration of ions into 
and out of existing silicate structures. The diffusion of Al 3+ or Mg 2+ out of the lattice 
was supposedly balanced by the inward diffusion of other ions. Such diffusion is un- 
likely. In mica, for example, a cation diffusing out of octahedral coordination would 
leave behind a void and many unsatisfied bonds. Such diffusion would be against an 
enormous gradient in electrical potential. The ion would also have to break through 
several tetrahedra to reach the lattice surface. The cation diffusing from the soil so- 
lution would be attracted by the electrical potential but would also have to break 
through the tetrahedra. Furthermore, the replacing cation would have to be similar 
in size and charge to the vacated cation. That the ion diffusing from the soil solu- 
tion would have the appropriate size is highly improbable. The common cations in 
soil solutions are Ca 2+ , Mg 2+ , Na+, and K + . Only Mg 2+ from this group fits into 
octahedral configuration. None of these cations normally occupies a tetrahedral posi- 
tion, and none would account for the differences in tetrahedral composition between 
mica and secondary minerals. Distortions during such ion diffusion would strain the 
crystal badly and probably cause its total rupture. The result would be more or less 
complete mineral breakdown before the ions recombined into a new mineral. 

Despite the unlikelihood of secondary mineral formation by ion substitution into 
or movement within an existing solid, some secondary 2:1 layer silicates apparently 
are formed by solid-phase changes of mica fragments inherited from the parent ma- 
terial. Hydrous mica, for example, is a product of chemical weathering as well as 
mechanical breakdown of mica. Hydrous mica, in turn, can be modified directly to 
vermiculite, montmorillonite, or chlorite. The process is not completely understood, 
but seemingly involves the outward diffusion of K + from between the layer lattices 
and a subsequent or simultaneous reduction of charge within the layer lattice. 



o 

D, 

E 



n 




co 




+■* 




0) 




O) 




C 




"35 




CO 




9> 


>*-a 


1_ 


w i/: 


o 
c 


'§"! 


•£ 


IS "> 


o 


=> .E 


0) 




•o 




h. 




o 




c 








<o 




■*-» 




c 




<1> 




E 




c 




o 




k 


«j 


'> 


1= 


c 


<U 


d) 


E 


•■■ 


c 


o 

<0 


c 

1- 


c 


> 

c 


*■* 


III 


c 




<d 




« 




01 




i_ 




Q. 




<A 




flj 




*- 




« 




C 








E 








15 


c/: 


2 


S3 


T3 
(A 


u 


P 


"« 


a— 


t. 


^-** 


<u 


JK 


c 


I. 




<0 


i 


C 




Q. 




c 




o 




E 




E 




o 




o 




<d 




t^ 




0) 


<0 


X) 

■5 


53 




'> 



£ 

x: 



o 



s 






- ft ~ P. ~ ~ 



o> o 

BO <" 



DO — 

2 ° 

~" 'Si 



M — 

2 o 



C 

o 
o .£ 

bo 3 *— . 
•S o -g 

y .e « 

•§ E -° 
x -i, -a 

Si § 

E ^ * 
— 12 o 

ca c ■• 

T3 



rt 



1) 



3 


B 

3 


00 


o 




C 


£ 


C 


« 


V- 


XI 


jy 


c 


<u 


o 


o 


M 


t-H 


X) 


u 




(D 


rt 


o 






VJ 

c 

3 



od 



o 



a; 



JU 

u, X> 

id rt 

O => 

z 



*0 nf 



,y HM ^ >"-* JW 



s 

■a a 
£ U 

V III 



id x <u x <u 

3 2 3 x 3 

o o o 

co </> en 



<N 

O o 

O O 
■u "■ 
!- t; 
a, 3 
o 



C 
o 

E 



as 



a: 



oi 



ai 



<0 



XI 

o 

E 8 



o 


o 


O 


g 

V, 

3 


E 

a 


o 

xl 


O 

c 


E 


& 

o 
E 


-5 


G 
O 





o "< t: -x: 

*e C ri -rt 



a. . 
E "o 

33 c ^ >-, g 
« s ? j= b 

X! o c; 

CO J P 



£D 



o 

XI 



U 






£ 



a. 



o 
Q 



as 







c3 


& 


«3 


V; 


t/i 


o 


3 


M 


3 


1 


O 


<J 


O 


0) 


O 


o 


a 


^-. 


a 


4> 


w 




00 


5? 



^3 S O 



X 
Cu 



S Q O 



2 ^ 



3 
>. O 



C 2L 
X3 ^3 

O ~ 



O 
i 



5 U. 
I- ^ 

3 B 



fti 



CS 


Vi 


t-t 


^ 


00 


U 


XI 


2 


op 




2 


- — 

IE 


■o 


e- 




o 

E 




?3 



o 



x 
a. 



•=+ -i 



,3 








^ 

3 




c ft 


eo 


>-. 






C 


Xj 


B 




c 


X 




X 


X 


'C 


oo <o 
SI 


§ 


O 

l-H 


u 


> 


o 


p 


xl 

o 


XI 


>1 

a. 


3 


O 


a. 


< 


s 





193 



a* o 

D .S 



tu 



U 



•D 
4> 

C 



o 
o 



* 



1 » 

8 | 

wo O 



^ O 

o o 

1 1 



'I 2 



B &£■* 



-a 

c 

C3 


o 
c 
© 


a 

a 

CO 

O 

3 


CO 

cu" 




XI 




c 


73 


'co 






z 


ccj 


O 

a. 


>-* 




O 

s 


ccf 


co 


c 
•C 
o 


"a 
x: 




CO 


u 

o 


CO 

o 


2> 


CO 

13 


4> 


3 


.c 


to 


'C 


3 


<u 


•a 


c3 


X) 


a> 


_u 


a 


2 
a. 




"5b 


6 


1 


o 










■i> 


</> 










> 



■a 



S « 



3 «< 



T3 

c 



M ig 



2 ■< 






3 .1= ix, 
53 2 4J m- 






8-3 8 

s S 3 

O o 

U co 



5SOUOU 



OS 




2 
a 






o o o o 

££££ 



o 

CO 



r * a 
*■ o- -a 



.V 



CO 
(-1 

o- ii '-5 tu 
J2 n o •, 



o 
■a 
'B, 
W 





a 




o 




C 




•a 
c 




>i 




CO 




o 








CS 




CO 




IS 




CU 






o 




e 




o 




S 


CO 


e 




i 




s 


CD 

•a 


X 

CD 








o 


o 

CD 
£ 
O 
O 


X> 


CO 

cu 

o 
.«> 


cd 

s 

X> 


co- 

13 

CD 


co 

3 

X) 

CO 

'5 


i 

a> 
to 

3 


<^4 




c 


CO 


CO 




o 

O 


co 

'o 

co 




CD 


'a 




3 


_c 


"u 


1 


>. 


'i/i 


CJ 

CD 


CD 


1 


t/1 


H 


u, 


a. 


co 


rJ 


2 


u 





3 



00 

o 
"o 



OS Oi 



op 

T3 



00 

T3 



o .S 
8 M 

g. a- 
» E 



U 



x: 

B- 
o 

E 



2 xi <y a 

- - •- x> 

XI 



£ 5 



CP A 



a a 

CO *-< 

3 " - 
O 

g -a 



J2 P 



U 



o 



i2 



x> 



N 



OS 



N 



c .- 

rt <o 

Is 

13 3 

u. g- 

P ^° 

o = 



13 '-» 

-3 to 



E 
c 
o 
.3 
S 



u 



4J 

E 

Z 



o 

vj O 

C 'S 

3 O 

o 73 

Sfc. o 

.a -8 

O .^ 

w -j 

<L> -5 

> O 

T3 en 

i> .a> 

3 •« 

— .£• 

o 13 

^ 13 

>, 2 

— <D 

JS X) 



3 

•a 
o 






■a 

s 
o 




£. 






o 




V 




£ 


;/> 




oo 




o 






t-< 






3 




_s 




1 

o 
a. 


X> 


</l 


c 

oo 
c 




c 
o 


c 
<u 

E 
c 
o 






00 


•3 


c 


3 






c 




2 


js> 


12 


o 
o 


> 

c 


c 
o 


2 
<3 
E 


2 


00 

c 
5 


00 
73 


"2 


o 


Otj 


x5 


3 


'S 


'£ 


d 


'3 


■o 


o 


o 




o 


c 






ri 








T3 

C 


1 


73 


c 

V> 


>> 


t3 


o 


<u 


E 


B. 


>-, 


o 


c 



•3 ID 

2 ?= 



1 



rs t *- H 



<L> 



« 



O T3 

O '3 



SSI 

II* 



•3 4) 

s"g 
>% <« 

2 <*- 
2 trf 

aj 
-a 
»-< 
o 
u 
oo 
e 



<S « 13 



u u a u u 



o 

CO 



a u. 



si 

00 








O 




2 


c. 


c 




Q« 




?~> 








<D 


nj 


<t£ 


> 


c 


<d 




a 


S3 


& 


E 


1 


o 

W5 





w 



a! 




aj Jj J> 

"2 , a .«* 
— S i: 

H oo co 



00 



OO 

c 



73 oo 
+ "'■= 
£g 

o g 

G <U 

° K 

« "C 

X P 

(C *> 

CJ "§) 

w 2 

oo [ti 
2 U 

&•& 



(Z o 



00 _ 

si 

U o - 

^ W -S 

U U c 

S.2 o - 
« .3 c 
o c ki 



oo 
2 



•a 

c 

<D 

a. 
•a 

X 
a, 
>, 

2 

op 
2 



■a x; 
oo oo 

2 2 



■a 

c 



X 
a. 



c 

XJ 

c 
a, 

4) 

•a 
i 

X 
tx 



2 



i 'r* "t"? "?* J— ? 



o -° 
Iff 



c oo c o 

O -3 O -3 

■3 „ -a E 

v> <D VI 

"3 oo -g >> 

2l B " 



a o 



a o 






% «e 



2 



■S — i 



o ~ 

j5 w: 



o 






o 








00 












o 




<u 


-3 








E 






^-^ 










3 












^ 


00'^ 


"3 






a 


x: 


x; 


>> >> 


o 


'jj 




2 




o 

'5. 
u 


? - 


u 

ID 


1 


> 


73 


< 


32 


> 


(*> 


E 


W 



o 



c ft 



w 



OS 



S3 
a. 



.S — 



o ■ 

o £; o 

<u ~ " 



6 .X 

S ^ 3 

«> 9 O 



T3 
O 



c; 



(3 



=3 =5 -a 

^ ^> '3 

^5 J3 /^ 

CL, Cl. O 



.-sis 
X o 



c 



195 



1 96 WEATHERING AND SOIL DEVELOPMENT 

Kaolinite, on the other hand, has no structural counterpart among the igneous 
minerals. It is also the most widespread of the crystalline clay mineral. The most 
likely mechanism for kaolinite formation is the complete breakdown of feldspar or 
mica particles and the precipitation of kaolinite from Al(OH>3 and Si(OH) 4 from the 
soil solution or from amorphous, less stable intermediates. 

The exact chemical conditions under which soil minerals form are not known at 
present. Our knowledge of the thermodynamic properties of soil and igneous min- 
erals is only sufficient to illustrate such conditions in a general way. In addition, the 
conditions apply only to equilibrium. Thermodynamics implies that only one energy 
state is the most stable state under a given set of conditions. The differences in en- 
ergy between some phases are so slight, however, that kinetic factors may be more 
important than energy differences in determining which minerals actually form. In 
addition, uncertainties in the available data on free energies of formation are some- 
times greater than the net free energies of potential reactions. Tabulated energies of 
formation for some soil minerals are little better than educated guesses. Free energies 
of formation also vary greatly with mineral chemical composition. 

Despite such reservations, the thermodynamic data now available provide a rea- 
sonably satisfactory picture of solution conditions under which well-defined miner- 
als form and decompose during weathering and soil development. The equilibrium 
between kaolinite and gibbsite, for example, can be written 

Al2Si205(OH)4+5H 2 = 2Al(OH) 3 + 2Si(OH) 4 (7.9) 

kaolinite gibbsite soluble silica 

The activities of the solids and of water are assumed to be unity. The equilibrium 
between the two minerals is therefore determined by the activity (concentration) of 
the only soluble species, Si(OH) 4 . The equilibrium constant for the reaction is 

K = (Si(OH) 4 ) 2 = 1(T 84 (7.10) 

The value of this constant comes from the free energies of formation of the reactants 
and products of Eq. 7.10. The equilibrium constant defines the soluble silica activity 
at which gibbsite and kaolinite are in equilibrium: 

(Si(OH) 4 ) = (l(r 8 - 4 ) l/2 = \<r i:i (7.11) 

At Si(OH) 4 activities less than 1() -4 2 , gibbsite is stable, kaolinite will not form, and 
any kaolinite present will decompose to gibbsite and soluble Si(OH) 4 . At Si(OH) 4 
activities greater than 10 -41 , kaolinite is the stable solid. Gibbsite is unstable at 
these concentrations and will react with Si(OH) 4 to form kaolinite. This discussion 
is continued in more detail in Appendix 7.1. 

All of the organic matter and part of the aluminosilicates, hydroxyoxides, and 
silica in soil exist in structures too small or too poorly crystalline to be detectable 
by x-ray diffraction. These amorphous materials are not well understood, but they 
should logically be among the most reactive of soil components, because their struc- 
ture is so open and their surface area so great. They represent a transition state be- 
tween unweathered parent materials and well-crystallized secondary soil minerals. 



MINERAL FORMATION IN SOILS 197 

Their high surface areas and low degrees of crystallinity suggest higher surface free 
energies and rapid transformation to more crystalline forms. Amorphous materials 
should tend to disappear from soils. 

Ions in soils, however, are in constant flux because of plant and microbial uptake 
and subsequent organic decay. This continual input of fresh amorphous material pre- 
vents attainment of equilibrium. The presence of many different ions in soil solutions 
hinders recrystallization of amorphous compounds by "poisoning" the surfaces of the 
growing crystal. Foreign adsorbed ions prevent more favored ions from contacting 
the surface and allowing the crystal to grow. Furthermore, the low solubility of soil 
minerals allows only slow ion movement through the solution phase to larger crys- 
tals of lower surface energy. Inorganic and organic coatings inhibit ion movement 
to and from the particles and partially neutralize the charges and unsatisfied bonds 
at mineral edges. Positively charged hydroxyoxide gels and negatively charged alu- 
minosilicates and organic gels can also neutralize each another. Such electrostatic 
attraction creates a greater resistance to weathering than if the phases remained sep- 
arate. 

Amorphous inorganic matter exists in all soils, but has been of greatest interest 
where it predominates in the clay fraction. Examples include soil formed from vol- 
canic ash in which high porosity, glassy mineral structures, and chemical instability 
permit rapid mineral weathering. When the resulting amorphous matter has an Si/Al 
mole ratio of about 1, it has been termed allophane. Definitions of allophane vary 
among investigators, as might be expected from the experimental difficulty of study- 
ing amorphous substances. 

The x-ray amorphous silica in soils includes opaline silica in the form of plant 
phytoliths. During plant growth, silica precipitates on the walls of plant cells. After 
death and organic decay of the plant, the silica phytoliths remain in the soil for many 
years as accurate representation of the cell wall. Phytoliths are visible under the 
microscope and can identify the plants in which they formed. 

7.4.1 Soil Carbonates 

In regions of limited rainfall, carbonates (particularly CaCCb) accumulate in soils. 
Where evapotranspiration exceeds precipitation, the downward flow of water through 
the soil profile is sufficient to remove only the most soluble weathering products, 
such as Na + salts. Intermittent rains can flush out soluble salts even when the amount 
of percolating water is 1% or less of the total rainfall. Less soluble compounds, on 
the other hand, accumulate because of limited water flow. The Mg 2+ and K + form 
secondary aluminosilicates. Secondary silicates containing Ca 2+ as a structural ion 
are rare, but Ca 2+ remains instead as an exchangeable cation and precipitates as cal- 
cite, aragonite, or vaterite (all CaC03) and occasionally as the more soluble gypsum 
(CaS04 • 2H?0). Calcite formed in soil allows little Mg 2 " 1 ' substitution into its struc- 
ture, and dolomite (CaMg(CC>3)2) apparently forms only under marine conditions. 

Calcite, aragonite, and vaterite can also accumulate in soils when hydrostatic pres- 
sure or capillary action move Ca 2+ - and C02-rich groundwaters upward in the soil 
profile. The loss of CO2 to the atmosphere and evapotranspiration of the water lead 



1 98 WEATHERING AND SOIL DEVELOPMENT 

to precipitation of CaC03. This mechanism sometimes accounts tor CaCC>3 accu- 
mulation in soils of more humid regions. 

The effect of chemical environment on CaCO.-? solubility in oceans and freshwa- 
ters has been considered at great length by geochemists, but under conditions that are 
not applicable to soils. They generally and tacitly assume (1) unlimited water con- 
tent, (2) constant Pqo 2 (where P is the partial pressure or concentration of CO2 in 
the gas phase, usually assumed to be that of the atmosphere, Pco 2 — 0.00035), and 
(3) that CaCCb is the only source of Ca 2+ . These assumptions are often invalid for 
soils. The water content of soils is limited and fluctuates. The CO2 concentration is 
closer to Pco t =0.01, increases to 0.2 in flooded soils and with root and microbial 
activity, and varies with the rate of upward diffusion of CO2 to the atmosphere. The 
Ca 2+ inputs from weathering and exchangeable Ca 2+ maintain relatively high Ca 2+ 
activities in the soil solution. 

Although the environmental conditions that bring about carbonate accumulation 
in soils are many and varied, the chemical reaction can be represented as simply 

Ca 2+ + H 2 + C0 2 = CaC0 3 + 2H + (7. 12) 

Alkaline conditions favor CaC03 accumulation, by consuming H + and driving the 
reaction to the right. Increasing Pqo 2 causes CaCC>3 to react further: 

CaCOj + C0 2 + H 2 = Ca 2 + + 2HCOJ (7. 13) 

so that CaCC>3 redissolves with increasing CO2 concentration in the gaseous phase. 
This is a condition of great interest to geochemistry. In soils, however, the relatively 
high Ca 2+ concentrations and limited water contents tend to force reaction 7.12 to 
completion and to repress reaction 7.13. The effect of Pco 2 > s therefore less impor- 
tant than in geochemical conditions, and CaCO} precipitates in soils despite the high 
Pcoi of soil air. In acid soils, CaCC>3 dissolves by reversing Eq. 7.12. 

When the mass of CaCC>3 in soils exceeds several percent, it controls both soil 
pH and soil solution Ca 2+ concentrations. Silicate reactions with H + and Ca 2+ in 
this case are relatively insignificant. The pH and Ca 2+ can then be calculated from 
Eqs. 7.12 and 7.13, the P C o 2 , and the solubility product of C&COy. 

(Ca 2+ )(CO^) = 10" 8 ' 4 (7.14) 

This model has been applied successfully to irrigated arid soils after the refine- 
ments of activity coefficients, presence of ion pairs (especially CaCO^), slight differ- 
ences in solubility products between calcite and aragonite, and the inhibitory effect 
of organic matter on carbonate precipitation are taken into account. The Pq^ is also 
a major chemical variable in soils. 



7.4.2 Carbon Dioxide 

Dissolution of carbon dioxide from the soil air into the soil solution affects the pH 
of the soil solution and the solubility of soil carbonates. Carbon dioxide dissolves 



MINERM. FORMATION IN SOILS 



199 



in water as C02(aq) and rapidly establishes equilibrium with water to form the weak 
acid H2CO3. The solubility of CO2 and other gases is governed by Henry's law, 
which relates the partial pressure (concentration in a gas mixture) of a gas to its 
concentration in aqueous solution: 



* H = M2i = l0 -,, 

Pco, 



(7. 1 5) 



Here H2CO3 refers to the sum of CO2, H2CO3, and C02aq dissolved in the aqueous 
solution. The H2CO3 dissociates successively to bicarbonate and carbonate ions as 
the pH increases: 



H2CO3 = H + + HCO3 
HCOJ = H+ + CO^ _ 



pK = 6.4 
pK = 10.3 



(7.16) 
(7.17) 



The resulting distribution of aqueous CO2 species with pH is shown in Fig. 7.4. 
The concentrations of the various ions can be calculated from these equations by 
assuming equilibrium with a specific Pco 2 - Rainwater is in equilibrium with the 
Pco 2 of 0.00035 in the atmosphere. In the absence of substances such as SO2, NO*, 
and NH3, the acidity of rainwater is fixed at pH 5.7, as calculated from Eqs. 7.15 and 
7.16. 

The solubility product of CaCOi in equilibrium with water and gaseous CO2 
yields an acidity of 



(H+) = (i0- 13 - 5 Pco,) 



1I/2 



(7.18) 



100% 



H 2 C0 3 



HCO3" 



CO3' 




FIGURE 7.4. Distribution of aqueous CO2 species with pH. 



200 WEATHERING AND SOIL DEVELOPMENT 

The pH of a solution in equilibrium with CaCC>3 and atmospheric CO2 is 8.3. Equa- 
tion 7. 18 yields a value of pH 8.5 because activity coefficient corrections are ignored 
and because of rounding-off errors in the exponents of the equilibrium constants. 

The pH of calcareous soils when measured in the laboratory is typically around 
8.3, if exchangeable sodium is low. Higher pH values indicate Na"'", and much more 
rarely K + , accumulation in the soil. Field pH values of calcareous soils are usually 
less than 8.3, however, because the CO2 concentration is higher in the soil's gas 
phase than in the atmosphere. Root respiration and microbial decay of organic matter 
release CO2, which must diffuse through soil pores to the atmosphere. The average 
Pco 2 i° the pores of agricultural soils is probably 0.003 to 0.03, 10 to 100 times that 
of the atmosphere. Many workers use the value of 0.01 as typical of CO2 in the gas 
phase of soils. A lower value might be better for warmer and drier soils. The actual 
CO2 concentration depends both on the rate of microbial and root respiration and on 
the diffusion rate of CO2 to the atmosphere. 

Gas diffusion is slower in wet and flooded soils, where soil pores are plugged by 
water. Gas diffusion in water is about l/10000th the rale of gas diffusion in air, or 
essentially nil in flooded soils where all soil pores are water-filled. The consumption 
of 1 mole of O2 during respiration yields approximately 1 mole CO2. In flooded 
soils, therefore, CO2 can almost completely replace O2 and reach a partial pressure 
of 0.2, equal to the value of O2 in atmospheric air. Since the gas volume in a flooded 
soil is minute, it is perhaps more instructive to say that the CO2 concentration in the 
soil solution is equivalent to Pco 2 = 0.2. At such concentrations, dissolved CO2 
has considerable influence on soil pH (Eq. 7.18). When soil solutions are extracted 
from soils, dissolved CO2 is slowly lost to the atmosphere. This causes large pH 
increases in extracts from alkaline and flooded soils, and the possible precipitation 
of CaC03 and of transition and heavy metal hydroxyoxides. The loss requires several 
hours so immediate measurements yield pH values more representative of actual soil 
conditions. 



7.4.3 Evaporites 

Although calcium carbonate formation in soil is a result of high evapotranspiration 
rates relative to precipitation rates, the term evaporite is usually restricted to com- 
pounds more soluble than CaCC>3. Where drainage water from surrounding soils 
accumulates and where the amount of percolated water is small compared to the 
amount of water evaporated, soluble salts tend to accumulate. This subject is dealt 
with in more detail in Chapter 11. The present section is restricted to the extreme 
case of natural salt flats and playas (former and intermittent lake beds). 

The high salt concentrations and accompanying high pH in these soils alter 
the course of soil mineral weathering toward the formation of minerals that are 
highly unstable under leaching conditions. The distinctive salts that form include 
the Na salts halite (NaCl) and trona (NaHC03), plus smaller amounts of sul- 
fates, borates, and similar salts of K and occasionally Li. Some secondary sili- 
cates also form, including the zeolite analcime (NaAISiiOc ■ 6H2O) and sepiolite 
(Mg4Si60is(OH)2-6H 2 0). 



STABILITY DIAGRAMS 201 

The formation of such minerals and authigenic feldspar under highly saline condi- 
tions is sometimes termed reversed weathering. This is appropriate in the sense that 
ions weathered in other locations are thus incorporated into new minerals instead of 
flowing to the sea. The minerals formed, however, are considerably different from 
the original igneous aluminosilicates, although their chemical compositions may be 
similar. 



APPENDIX 7.1 STABILITY DIAGRAMS 

The stability of minerals contacting aqueous solutions depends on the composition 
of the solution. Figure 7.5 shows the stability regions of major soil minerals at vary- 
ing concentrations of Si(OH) 4 , Ca 2+ , Na + , and H + . The soil-formed minerals— 
gibbsile, kaolinite, and montmorillonite — are stable at lower Ca activity and lower 
pH than the igneous Ca feldspar, anorthite. Gibbsite is stable at the lowest Si(OH) 4 
activity and montmorillonite is stable at the highest Si(OH) 4 activity. The stability 
region of Ca-montmorillonite is greater than that of Na-montmorillonite. In soils the 
concentrations of three aqueous species — Ca, H, and Si(OH) 4 — are limited but the 
concentration of Na in the soil solution is unlimited. Precipitation of CaC0 3 prevents 
the value of pH - ^pCa from increasing enough to precipitate anorthite. Anorthite is 
unstable under soil conditions and weathers to the soil-formed minerals. The upper 
limit of Si(OH) 4 activity is determined by the solubility of amorphous silica, which 
precipitates at (Si(OH) 4 ) = 10" 2S . 




FIGURE 7.5. Stability diagram o1 the Ca-Na-Ai-Si-0-H 2 system at equilibrium at 25° C. The 
Ca system is the front vertical plane, Na is the back plane. Braces denote moles of ion charge 
in solution. (From H. L. Bohn and Kittrich. 1984. Chem. Geol. 43:181.) 



202 WEATHERING AND SOIL DEVELOPMENT 

Figure 7.5 is approximate because the AG values of formation of Ca- and Na- 
montmorillonites are educated guesses. Figure 7.5 also ignores the slow kinetics of 
silicate reactions. In addition, the activities of Mg 2+ , K + , and Na + affect the stability 
of other soil minerals and their igneous minerals, which directly and indirectly affect 
the stability of Ca minerals. 

Mineral stabilities often depend on the activities of H + , other cations, and the 
Si(OH) 4 activity. The H + activity is usually inversely related to the activities of K + , 
Na + , Ca 2+ , orMg 2+ . The M + /H + ratio or pH — pM, although awkward to visualize, 
permits plotting both activities on one axis. High values of pH— pK and of pSi(OH) 4 
represent alkaline soils with little leaching. Low values of pH — pK and pSi(OH) 4 
represent acid soils and extensive leaching. 

The gibbsite-kaolinite boundary in Fig. 7.5 is at a Si(OH) 4 activity of 10 -42 . 
Kaolinite, therefore, will not form unless the soluble silica concentration exceeds this 
level. Any kaolinite present at lower silica concentrations will eventually dissolve to 
form gibbsite and soluble Si(OH) 4 . 

At higher Si(OH) 4 activities, Fig. 7.5 shows that kaolinite is unstable and is trans- 
formed to montmorillonite. Assuming that Ca 2+ is the only exchangeable cation in 
the system, the equation for the equilibrium between kaolinite and montmorillonite 
can be written as 

7AI 2 Si20 5 (OH)4+8Si(OH)4 + Ca 2+ = Ca(Al 7 Si| |0 3 o(OH) 6 ) 2 + 23H 2 + 2H+ 
kaolinite soluble silica montmorillonite (7.19) 

where montmorillonite for simplicity is given an idealized composition. The equilib- 
rium constant of Eq. 7.19 is 



(H+) 2 
(Si(OH) 4 ) 8 (Ca 2+ ) 



^ = , Ci ,o„v^H- (? - 20) 



The value of K eq can be calculated from the energies of formation of the components 
of Eq. 7.19. The equilibrium between kaolinite and montmorillonite depends on the 
H + , Ca 2+ , and Si(OH) 4 activities in solution. Rearranging Eq. 7.20 yields 

pH - ipCa = 4pSi(OH) 4 - \ log K eq (7.21) 

The equilibrium between kaolinite and montmorillonite is plotted according to 
Eq. 7.21 in Fig. 7.5. Montmorillonite is stable at high Si(OH) 4 activities and moder- 
ate Ca 2+ /H + ratios. 

The K and Na stability diagrams are similar, but the values of M/H vary. The Ca 
feldspars are the most unstable, and the K feldspars are the most stable, with respect 
to weathering. At higher M/H ratios (more basic solutions) other minerals are stable 
and would precipitate before the igneous feldspars, but free energy data for other 
silicates are lacking. 

Stability diagrams such as Fig. 7.5 should not be taken too literally. Free energy 
data for most minerals are uncertain. Even small errors are magnified by calculating 
the equilibrium constant, which is the antilogarithm of a small difference between 
large AG's of formation. In addition, the activities of the solid phases are assumed 



STABILITY DIAGRAMS 



203 



to be unity, i.e, the minerals are assumed to be pure. This is probably approximately 
true for the major minerals and for minerals that do not form solid solutions. The 
activities of the minor components of solid solutions are not known with certainty 
but are almost surely not unity. At present, stability diagrams serve as inexact but 
useful illustrations of the relations between minerals in nature. 

Occasionally the course of weathering can "reverse" in the sense that feldspar 
weathering will create secondary silicates that are unstable under the leaching and 
weathering conditions of well drained soils. Such reversal occurs because of the ac- 
cumulation of K + , Na + , Ca 2+ , Mg 2+ , and Si(OH) 4 in arid and poorly drained soil 
solutions. These secondary silicates include zeolites, evaporites, and the authigenic 
feldspars. The area denoted as soil solution in Fig. 7.6 shows the extreme concen- 
trations that have been reported in soil solutions. Within this range, several silicate 
minerals are stable. 



- o 



-" INT£RSTITlfli_ 
■ SOLUTIONS 
I 




-log Si (0H) 4 , or p Si(0H) 4 



FIGURE 7.6. Trie K-AI-St-0-H 2 equilibrium system at 25° C, superimposed on the ranges 
of compositions of soil solutions, oceans, and interstitial solutions. (From H. C. Helgeson, 
T. H. Brown, and R. H. Leeper. 1969. Handbook of Theoretical Activity Diagrams Depicting 
Chemical Equilibrium in Geologic Systems Involving an Aqueous Solution at One Atm and 0° to 
300° C. Freeman & Cooper, San Francisco; and S. V. Mattigod. 1976. Thesis. Washington State 
University, Pullman, WA.) 



204 WEATHERING AND SOIL DEVELOPMENT 

At 0° C, the secondary minerals gibbsite, kaolinite, and montmorillonite are more 
stable relative to the igneous feldspars and mica than at 25° C. The stability fields 
of gibbsite, kaolinite, and montmorillonite are appreciably larger at 0° C. Because 
soil temperatures lie between and 25° C, Fig. 7.5 rather conservatively illustrates 
the stability of soil-formed minerals. Assembling Figure 7.5 requires comparing the 
stabilities of many likely minerals and eliminating those that are unstable relative to 
other minerals. Muscovite mica is thereby found to be stable at 25° C only within a 
narrow region of solution compositions between the stability fields of microcline, a 
mineral identical in composition to orthoclase feldspar, and orfhoclase. At 0° C, the 
stability region of muscovite disappears. 

Figure 7.6 shows the solubilities of quartz and amorphous silica in relation to 
the minerals of Fig. 7.5. The solubilities of substances having the empirical formula 
Si02 or S1O2 • «H20 have been studied for decades. These studies are much more 
complicated than they appear, because of the reluctance of soluble silica to reach 
equilibrium or even metastable equilibrium with its solid phases. Soluble silica tends 
to polymerize slowly in supersaturated solutions rather than to precipitate cleanly. 
In addition, the solid phase that precipitates is often amorphous silica instead of the 
most stable phase, quartz or its close relative chert. Amorphous silica is metastable 
and much more soluble than quartz. The solubilities of amorphous silica and quartz 
are often assumed to be the upper and lower limits of silica solubility in soils. Viewed 
from the range of soil solution compositions shown in Fig. 7.6, silica concentrations 
can be less than the equilibrium solubility of quartz even though quartz is almost al- 
ways present in the sand fraction of soils. The slow kinetics of silica reactions and the 
slow release of Si(OH)4 during weathering create wide deviations from equilibrium. 

Figure 7.6 also shows the concentration ranges of ocean waters and of the intersti- 
tial water in minerals. The composition of interstitial water is not well known, but the 
region shown is a reasonable guess. Ocean and interstitial waters are as influential 
as the soil solution in affecting mineral transformations at the earth's surface. The 
differences in composition of these three solutions also give rise to wide variations 
in the nature and distribution of soil clay minerals. 

BIBLIOGRAPHY 

Feldman, S. B., and L. W. Zelazny. 1998. The chemistry of soil minerals. In Future Prospects 
for Soil Chemistry (P. M. Huang, D. L. Sparks, and S. A. Boyd, eds.). Soil Science Special 
Publ. 55, American Society of Agronomy, Madison, WI, pp. 139-152. 

Jackson, M. L., and G. D. Sherman. 1953. Chemical weathering of minerals in soils. In Ad- 
vances in Agronomy, vol. 5 (A. G. Norman, ed.). Academic, New York, pp. 221-317. 

Jenny, H. 1941. Factors of Soil Formation. McGraw-Hill, New York. 

QUESTIONS AND PROBLEMS 

1. Show how to calculate the pH of rainwater, assuming equilibrium with CO2 and 
the absence of other acidic or basic solutes. 



QUESTIONS AND PROBLEMS 205 



2. Derive Eq. 7.18. 



3. Describe the major minerals and states of calcium, magnesium, potassium, and 
sodium as they cycle from rock to soil to sea to rock. 

4. From Tables 7.2 and 7.4, calculate the average residence times of calcium, mag- 
nesium, and potassium in soils. 

5. Show the equations for the equilibria between anorthite and gibbsite kaolinite 
and calcium montmorillonite. 

6. If the Na" 1 " residence time in the oceans is about 10 8 years, what is the Na' 1 
residence time in soils? What assumptions are necessary? How many times on 
the average has Na + recycled from rock to soil to oceans? 

7. For Ca 2 ' 1 ', Mg 2+ , K + , P 5+ , Si 4+ , or other selected chemical elements, trace all 
the possible pathways that the ion might follow from its state in igneous rock 
through soils to its most stable state at the earth's surface. 

8. Explain why soil feldspar panicles tend to be larger than soil kaolinite particles 
and why both tend to be larger than montmorillonite particles? 

9. Calculate the Ca 2+ activity in equilibrium with CaCOs and Pco 2 = 0.2 and 
0.0 1 at (a) pH 8.0 and (b) pH 5.0. 



8 



CATION RETENTION 
(EXCHANGE) IN SOILS 



Probably the most important and distinctive property of soils is that they can retain 
ions and release them slowly to the soil solution and to plants. The retention prevents 
concentrations that are too high and too low. The evolution of plants has taken ad- 
vantage of this buffered range of ion concentrations that soils make available in the 
soil solution. Over most of the earth's surface, the availability of these ions in the 
soil solution is adequate, but not necessarily ideal, for plants. Crop and horticultural 
plants and a desire for maximum yield place greater demands on the soil and may 
require adjusting the native soil solution. Adjustments by fertilization, liming, and 
salt removal are usually temporary. The soil and climate tend to return the soil to its 
native state. 

Ion retention is actually ion exchange. Soils give up other ions, H + or OH~ and 
HCOJ, in equal amounts to those retained. When trace ions are removed from the 
soil solution, the ion exchange to the soil solution is often unnoticed. The retention 
of organic, nonionic substances usually results in their degradation by soil microbes 
and conversion to CO2 and water. This chapter is concerned with the exchange, the 
retention and release, of cations between soil particles and the soil solution. 

Soil chemistry has stressed cation retention and exchange and has almost ignored 
anion retention and exchange. This unfortunate bias is because the clay particles of 
most soils of Europe and North America have a net negative change. The amount 
of cation exchange is therefore greater than anion exchange. Had soil chemistry be- 
gun in Australia, in soils of volcanic parent material, or in highly weathered tropical 
soils the bias might be toward anion exchange and retention. Soils have both a neg- 
ative charge that retains cations and a positive charge that retains anions. We usually 
measure the soil's net charge and that is usually negative. 

Cation retention by soils can be roughly divided into the weaker electrostatic inter- 
action of soil particles with the alkali and alkaline earth cations and the soil's stronger 

206 



ELECTROSTATIC CATION RETENTION (CATION EXCHANGE) 207 

chemical bonding with trivalent and transition metal cations. Chemical bonding is the 
interaction of polyvalent cations with O 2- " and OH - ligands of aluminosilicates, hy- 
droxyoxides, and phosphates, plus retention of weak Lewis acids by soil organic mat- 
ter. Chemical bonding, also called precipitation or strong adsorption, is discussed in 
Chapter 3 and is related to the dissolution-precipitation reactions of classical chem- 
istry. The weaker, electrostatic retention of ions is distinctive to soils and colloidal 
systems and creates the major reservoir of the essential macroelement cations for 
plants and all living organisms. Some generalizations can be made about the attrac- 
tion, exchange, and retention of cations by soils: 

1. Relatively weak (electrostatic) attraction— alkali and alkaline earth cations 
(mainly Ca, Mg, K, and Na) 

Nonspecific, depends mostly on the concentration ratios on the solid vs. 
the soil solution and on the ion charge ratio. Some clay minerals prefer 
one ion over others. 

Reactions are fast and reversible; time scale is seconds and minutes. 

Amount of retention depends on soil's cation exchange capacity, the neg- 
ative charge of soil particles. 

Largely due to aluminosilicate clay minerals plus soil organic matter. 

2. Strong (chemical bonding) attraction — H, Al, Be, Ti, transition metal, and 
"heavy metal" cations 

Specific, that is, the strength of attraction depends mostly on the cation's 
water solubility and the amount of that cation on the surfaces of soil parti- 
cles. 

Reaction time scale is rapid at first, but continues at ever-slower rates for 
long periods. 

Amount of retention depends on soil pH rather than on the charge proper- 
ties of soil clays. 

Aluminosilicates are less important, Fe and Mn oxides are more impor- 
tant, in this retention than in electrostatic cation retention. The retention 
is generally much stronger than that predicted by aqueous solubility prod- 
ucts. 

Mechanisms of retention are complex and grade gradually from one to 
another. 

Organic matter increases the range of sorptivity, possibly by adding soft 
Lewis base character. 
Amorphous materials retain more than crystalline. 



8.1 ELECTROSTATIC CATION RETENTION (CATION EXCHANGE) 

Exchangeable ions are those ions replaced by neutral salt solutions flowing through 
soils. Soluble salts are removed by water alone. Salt solutions also exchange anions 



208 CATION RETENTION (EXCHANGE) IN SOILS 

from soils, but because soil colloids are mostly negatively charged, generally more 
cations exchange than anions. The major exchangeable cations are, in order of de- 
creasing amounts, Ca 2+ , Mg 2+ , K + , and Na + . The retention helps prevent leaching 
losses during weathering; plants exchange for these cations and absorb them by re- 
leasing H + . 

In humid and temperate region soils, most of the plant-available alkaline earth 
and alkali metal cations neutralize the soil's negative charge. A typical agricultural 
loam soil contains about 20 000-30 000 kg ha~ ' of exchangeable cations in its root 
zone (0.5 m depth). Roughly 80% is Ca, 15% is Mg, 4% is K, and 1% is Na. As soil 
acidity increases, Al 34 " and H + are also exchangeable on soil surfaces. Arid soils may 
contain salts considerably in excess of the charge-neutralizing cations. Exchangeable 
cations and soil solution salts are very important to plant productivity and are easily 
manipulated by liming, irrigation, and fertilization. Hence, cation exchange has long 
been an important part of soil chemistry research. 

Thompson and Way conducted the first recorded studies of cation exchange in 
Rothamsted, England, in 1850. They showed that passing an ammonium sulfate so- 
lution through soil columns leached calcium sulfate out of the soil. The predominant 
cation in the aqueous solution had changed from ammonium to calcium because 
of cation exchange in the soil. Thompson and Way showed that the exchange was 
very fast and reversible, and that the amount of ammonium ion retained equaled the 
amount of calcium released. Subsequent work has refined and supported these find- 
ings and has measured the cation exchange capacities of soil and soil components, 
the relative affinities of soils and their components for various cations, and the effects 
of changing soil pH on exchange reactions. 

Cations in soils are roughly in three major categories: solid phase, exchangeable, 
and soluble. Weathering and organic decay release cations that vary in charge, size, 
and polarizability, so they respond differently to the soil surfaces and other ions they 
encounter in the soil solution. Small, polyvalent ions tend to reprecipitate/adsorb in 
soils by forming strong chemical bonds with aluminosilicate and hydroxyoxide sur- 
faces. Larger, lower-charge cations (mainly Ca 2+ and Mg 2+ ) instead associate more 
weakly with surfaces of the solid phase and are the exchangeable ions. The largest, 
lowest-charge cations (K + and Na + ) are weaker competitors for surface charge neu- 
tralization and tend to dominate in the bulk soil solution away from the charged sur- 
faces (the soluble ions). Weathering tends to remove the soluble and exchangeable 
ions from soils. 

The distribution of major exchangeable cations in productive agricultural soils is 
generally Ca 2+ > Mg 2+ > K + > NH^" «s Na + . The composition of the exchange- 
able cations in different soils tends to be much more uniform than the composition of 
the parent material rocks from which the soils are derived. The general effect of soil 
reactions is to smooth out the differences between soil parent materials and inputs, 
and to create a relatively uniform distribution of exchangeable ions for plant growth. 

The soil-forming factors can modify the distribution of exchangeable cations from 
this desired state. Table 8. 1 shows examples of exchangeable cations found in a wide 
variety of soils. The Merced soil has high exchangeable Na because it is poorly 
drained in an arid climate, it has no drainage to the sea, and upwelling water from a 



ELECTROSTATIC CATION RETENTION (CATION EXCHANGE) 209 

Table 8.1 . CEC values and major exchangeable cations of selected soils* 

CEC Exchangeable Cations (% of Total) 

Soils pH (mmolkg-') Ca 2+ Mg 2+ K + Na+ H + (Al 3+ )' ; 

Average of agricultural 7.0 383 79.0 13.0 2.0 6.0 — 

soils (Netherlands) 
Average of agricultural 7.0 203 65.6 26.3 5.5 2.6 — 

soils (California) 
Chernozem or Mollisoll 7.0 561 84.3 11.0 L.6 3.0 — 

(Russia) 
Sodic Merced soil 10.0 189 0.0 0.0 5.0 95.0 0.0 

(California) 
Lanna soil, unlimcd 4.6 173 48.0 15.7 1.8 0.9 33.6 

(Sweden) 
Lanna soiljimed 5.9 200 69.6 I I.I 1.5 0.5 17.3 

(Sweden) 

"From F. B. Bear, cd. 1964. Chemistry in the Soil, 2d ed. American Chemical Society, Washington, DC, 
p. 167. 

''Probably includes some titratablc acidity (Chapter 10.) 



high water table evaporates at the surface. Under those conditions Ca precipitates as 
CaC03, Mg, and K can form secondary soil minerals, leaving Na as the major cation. 
Exchangeable Na may exceed K in the Netherlands soils because of atmospheric in- 
puts of NaCl from the nearby ocean. The high Mg 2+ content of the California soils 
may reflect the high-Mg content of rocks found in volcanic and geologically ac- 
tive regions. Exchangeable Al 3+ is present in appreciable quantities in acid soils 
(pH < 5.5), such as the Lanna soil in Sweden. This soil is formed from granitic 
rocks in dense forest under conditions of slowly weatherable parent material, high 
rainfall, good drainage, and organic acids from organic matter decomposition. This 
strongly acid soil (pH < 4) contains considerable exchangeable Al and some ex- 
changeable H + . 

Productive agricultural soils are characteristically Ca dominated. One important 
part of sustainable agriculture is to maintain Ca dominance. Despite the wide range 
of soil-forming factors in Table 8.1, Ca is the predominant cation in all but the ex- 
tremely sodic (and barren) Merced soil. 

The sum of exchangeable Ca a+ , Mg^ + , K + , Na + , and Al i+ generally equals, for 
practical purposes, the soil's cation exchange capacity (CEC). The CEC varies from 
1 mmol(-f) kg - ' for coarse-textured soils to 500 to 600 mmol(+) kg - ' for fine- 
textured soils containing large amounts of 2:1 layer silicate minerals and organic 
matter. 

In 1 850, Thompson and Way found cation exchange to be reversible, stoichio- 
metric (the amount released, as moles of ion charge, equals the amount retained), 
and rapid. Since then some refinements have been studied and some exceptions have 
been found, but their results are still generally valid. Although cations are preferred 



210 CATION RETENTION (EXCHANGE) IN SOILS 

in varying degrees by soil colloids, even strongly adsorbed cations can normally be 
replaced by manipulating solution conditions. An exception to this generalization 
of reversibility is the preferential retention of many polyvalent cations (especially 
trace metals, weak Lewis acids) by soil organic matter. Such cations, which are 
thought to be partially covalently bonded, can be displaced only by other polyvalent 
cations capable of forming even stronger covalent bonds. Other exceptions include 
cation fixation reactions, described later in this chapter; cases where large organic 
cations, such as the pesticides paraquat and diquat, are physically prevented (steric 
hindrance) from approaching certain interlayer exchange sites; and cases where mul- 
tivalent cations are preferentially adsorbed because they can simultaneously balance 
several closely adjacent exchange sites. 

Because cation exchange reactions are stoichiometric, the sum of all exchangeable 
cations present at a given pH and CEC varies little or not at all with cation species. 
For example, consider the exchange reaction 

CaX + 2NH+ = (NH4) 2 X + Ca 2+ (8. 1) 

where X designates a cation exchanger. Two ammonium ions replace one calcium ion 
to preserve the stoichiometry of the reaction. Exchangeable cation composition and 
CEC values normally are expressed as moles of ion charge kg - ' , formerly meq/100 g 
(milliequivalents per 100 g). 

Exchange reactions are also rapid. The exchange step itself is virtually instanta- 
neous. The rate-limiting step often is ion diffusion to or from the colloid surface. This 
is particularly true under field conditions, where ions may have to move through tor- 
tuous pores or through relatively thick, stagnant water films on soil colloid surfaces 
to reach an exchange site. The need for diffusion can produce hysteresis (the extent 
or speed of reaction depends on direction of the reaction) for some ion exchange re- 
actions. Under laboratory conditions, samples normally are shaken during exchange 
reactions, to speed ion movement and to minimize the thickness of stagnant water 
layers on soil particle surfaces. 

Because of their reversibility, cation exchange reactions can be driven forward or 
reverse by manipulating the relative concentrations of reactants and products. In the 
laboratory, common techniques for driving the reactions toward completion are to 
use high (> 1 M) concentrations of exchanging cations and to maintain low concen- 
trations of product cations by leaching or repeated washings: 

CaX + 2Na + (high concentration) = Na 2 X + Ca 2+ (low concentration) (8.2) 

to form insoluble precipitates 

CaX + Na 2 C0 3 = Na 2 X + CaC0 3 (precipitate) (8.3) 

or to form volatile gases 

NH 4 X + NaOH = NaX + NH4OH = NaX + H 2 + NH 3 (gas) (8.4) 



ELECTROSTATIC CATION RETENTION (CATION EXCHANGE) 21 1 

For exchange between cations of differing charge, diluting the solution favors reten- 
tion of the more highly charged cation. For example, Eq. 8.1 has a reaction coeffi- 
cient k: 

[(NH 4 )2Xl[Ca 2+ l 

k — J-- — (8.5) 

[CaX]lNH+l 2 

where brackets indicate concentrations (mol L _l or mol kg -1 ) rather than activities. 
Rearranging Eq. 8.5 gives a typical cation exchange equation: 

[(NH 4 ) 2 X] JNH+J 2 

= k £- — (8.6) 

LCaX] [Ca 2+ ] k ' 

Because of the squared term on the right of Eq. 8.6, the ratio of ammonium to calcium 
in the colloid's double layer changes with total, as well as relative, salt concentra- 
tion of the bulk solution. This dependence of cation exchange on cation valence has 
been termed the valence dilution effect. As an example, consider a solution having 
[NH+] = [Ca 2+ ] = 1 mmol L - ' . The ratio [NH+] 2 /[Ca 2+ ] in this case equals l 2 / 1 , 
or 1 mmol L -1 . Upon tenfold dilution, the ratio is [0. 1] 2 /[0.1], or 0.1 mmol L _1 . 
Hence, the ratio of ammonium to calcium on the colloid decreases with dilution (Ta- 
ble 8.2). The total quantities, but not the concentrations, of replacing ions remained 
constant. Table 8.2 shows the absence of a dilution effect for cations of the same 
valence. The percentage of calcium replaced by barium remained virtually constant, 
but that replaced by ammonium decreased with decreasing salt concentration. 

Exchanging one cation for another in the presence of a third (complementary) 
cation also becomes easier as the retention strength of the third cation increases. 
For example, replacing calcium by ammonium is easier from a Ca 2+ -AI 3+ soil than 
from Ca 2+ -Na + soil. The fraction of the CEC satisfied by the tightly bound Al is in 
effect blocked off, and the Ca and ammonium ions compete for a smaller number of 
exchange sites. 



Table 6.2. Replacement of exchangeable calcium from 1 mmol 
of montmorillonlte exchange sites by a constant amount 
(1 mmole of ion charge) of barium or ammonium, at varying 
replaclng-cation concentrations 9 

Percent Ca 2+ 
Solution Added Replaced by: 



liters 


mol charge L ' 


Ba 2 + 


NH+ 


0.025 


0.04 


49.7 


29.8 


O.I 00 


0.01 


50.2 


20.8 


0.200 


0.005 


50.8 


16.6 


0.400 


0.0025 


52.7 


15.2 



"Adapted from P. Schachtschabel. 1940. Kolloid-Beihefte. 51:199-276. 



212 CATION RETENTION (EXCHANGE) IN SOILS 

8.1.1 Exchange Selectivity 

The attraction of cations for negatively charged colloid surfaces is qualitatively 
described by electrostatic attraction and repulsion, following Coulomb's law (Ap- 
pendix 8.1). A major limitation of this simple electrostatic approach, however, is 
its failure to predict differences in preference or selectivity of colloid surfaces for 
cations of the same valence. Such preference is related to the relative hydrated sizes 
or to the relative energies of hydration of the various cations. Ions of smaller dehy- 
drated radius have a greater density of charge per unit volume. Hence, they attract 
waters of hydration more strongly and have a larger hydrated radius. An ion of larger 
hydrated radius is held less tightly by coulombic attraction. Partially dehydrated ions 
can approach the surfaces more closely and generally are retained quite tightly by 
soil colloid particles. 

These generalizations arise from data such as in Table 8.3. The data were gener- 
ated by saturating a montmorillonite suspension with a given ion and then measuring 
the quantity of that ion released when a symmetry (amount equal to the CEC) of 
either NH 4 C1 or KC1 was added. 

The most important factor determining the relative extent of adsorption or desorp- 
tion of a given ion is its valence. Divalent ions in general are retained more strongly 
than monovalent ions, trivalent ions are retained even more strongly, and quadriva- 
lent ions such as thorium Th 4+ are essentially unreplaced by an equivalent amount 
of KC1. 

Within a given valence series, the degree of replaceability of an ion decreases as 
its dehydrated radius increases. An apparent exception is the "H + " ion. Monovalent 

Table 8.3. Relation of ion charge and ion size to Ion retention 3 





Crystallographic 


% Released by 


Ion 


(Dehydrated) Radius (nm) 


NHJ" or K+ 


Li+ 


0.068 


68 


Na+ 


0.097 


67 


K+ 


0.133 


49 


NH+ 


0.143 


50 


Rb H 


0.147 


37 


Cs+ 


0.167 


31 


"H+" (Al 3+ ) 


(?) 


15 


Mg 2+ 


0.066 


31 


Ca 2 + 


0.099 


29 


Sr 2 + 


0.112 


26 


Ba 2+ 


0.134 


27 


Al 3+ 


0.051 


15 


La 3 + 


0.102 


14 


Th 4+ 


0.102 


2 



"Modit'ed from H. Jenny and R. F. Rcitcmcier. 1935. Reprinted with permis- 
sion from J. Phys. Chem. 39:593-604. Copyright by the American Chemi- 
cal Society. 



ELECTROSTATIC CATION RETENTION (CATION EXCHANGE) 213 

"hydrogen" in this case behaves more like trivalent lanthanum. Work with acid soils 
and clays since the early 1 950s has demonstrated that "hydrogen" clays are unstable 
and rapidly decompose to produce aluminium-saturated clays. Hence, the "H + " entry 
of Table 8.3 probably represents Al 3+ . 

Relative ion replaceability, or ease of removal from specific colloids, has been 
called the lyotropic series. For example, the data of Table 8.3 could be written as 

Li + « Na + > K + % NH+ > Rb + > Cs + « Mg 2+ > Ca 2+ > Sr 2 * 
ss Ba 2+ > La 3+ % H(Al 3+ ) > Th 4+ 

in order of increasing strength of retention by montmorillonite. The order of the 
lyotropic series is explainable if the cations at the colloid surface include a layer 
of specifically adsorbed or partially dehydrated cations, the so-called Stern layer 
(Appendix 8.1). The composition of the Stern layer can be estimated from coulombic 
calculations if individual ion characteristics (e.g., hydrated radius and polarizability) 
are considered. 

Soil colloids of high charge density, that is, of high charge or CEC per unit surface 
area, generally have the greatest preference for highly charged cations. For example, 
vermiculite normally retains more Ca than does montmorillonite from a mixed Na '"- 
Ca 2+ solution. Hence montmorillonite has a higher exchangeable Na percentage than 
vermiculite at the same bulk-solution Na and Ca concentrations. The monovalent 
cations NH^ and K + are often exceptions to this generalization, because of their 
unusually strong preference by mica and vermiculite (discussed in greater detail be- 
low). Partially covalent bonding and/or complex formation may contribute to a sim- 
ilar preference of soils high in organic matter for many polyvalent cations. Raising 
soil pH can also change cation selectivity by increasing soil CEC and thus increasing 
the preference for polyvalent versus monovalent ions. 

In addition to coulombic preferences related to ion size, certain colloids exhibit 
unusually high preferences for specific cations. An example is the high exchangeable 
magnesium content of vermiculite. Hydrated Mg 2+ apparently fits so well into the 
water network between partially expanded sheets of vermiculite that Mg is preferred 
over a wide concentration range (Fig. 8.1). The dashed line in the figure indicates no 
ion preference by the colloid, and the dotted line shows a more typical case of Ca 2+ - 
Mg 2+ exchange with an exchange coefficient of 1.5. At low Mg 2+ concentrations, 
vermiculite prefers Ca over Mg because the hydrated Mg 2+ is larger than hydrated 
Ca 2+ . As soon as enough Mg (>40%) is present in solution to exert a significant 
effect on the interlattice water network, the curve shifts to a pronounced preference 
for Mg. Although normal soil solutions have relatively high Ca 2+ /Mg 2+ ratios, the 
crossover on the figure occurs at Ca 2+ /Mg 2 "'" ratios that are attainable under some 
natural conditions. These conditions can occur when calcium carbonate is precipi- 
tating, when former marine sediments are contributing soluble salts, or when high 
Mg micas are weathering to vermiculite. Vermiculite then is an excellent scavenger 
of Mg ions and becomes nearly saturated with Mg even when exposed to apprecia- 
ble Ca concentrations and/or monovalent cations. To replace the Mg from natural 



214 



CATION RETENTION (EXCHANGE) IN SOILS 











Ca IN SOLUTION 






I00 


00 


80 




60 


40 20 


C 


o 




I 




I 


L— °T~ 


/•' 
















<■' 














/^ • 
















• • 




~„ o 


80 


— 








• .-" 


— 


20 <_> 


CT 










/ .• 






2 










/ ■■ 




_J 


UJ 

a! 60 








la 


/ .• 
s 


- 


CD 

40 2 


< 










/ •• 




o 


UJ 










/ ..■ 




z 


o 








I / 






< 


Z 40 


- 










- 


60 J 


x 








/[ .-' 






X 


u 






/ 








UJ 


X 






/ . 










uj 20 




/ 










80 5 s 




/. 


i 




1 


I 1 




inn 





20 




40 


60 80 


100 










Mg IN SOLUTION 







FIGURE 8.1. Ca-Mg exchange in a vermiculite suspension. The clashed line represents an 
exchange constant ot 1 .0 (no preference); dotted line represents a Ca-Mg exchange constant 
of 1.5. (From F. F. Peterson et. al. 1965. Soil Sci. Soc. Am. Proc. 29:327.) 



vermiculites, the mineral must be repeatedly leached with high concentrations of a 
replacing ion. This procedure lowers the relative Mg concentration below the point 
of preferential adsorption. 

Another case of high preference for a particular ion is the preference of vermi- 
culites and of weathered edges of trioctahedral micas for K and similar exchangeable 
cations. Mica weathers to a vermiculite-like mineral with a decrease in layer charge 
accompanying the weathering process. Traditionally, the preferential adsorption of 
K, NH 4 , cesium, and rubidium by such minerals has been attributed to the excellent 
fit of the ions in the hexagonal or ditrigona] holes on vermiculite surfaces. The pro- 
cess is believed to be activated by the dehydration of large, weakly hydrated ions as 
adjacent silicate sheets approach one another during thermal motion or drying. 

Alternatively, the affinity has been explained by the relative hydration energies 
of various ions, plus the relative hydration energies of individual cation exchange 
sites on different minerals. The relatively small hydration energies of K, NH4, Rb, 
and Cs result in easy dehydration and strong retention. The hydration energy theory 
explains how Ba 2+ , with essentially the same crystal lographic radius as NH4, is not 
fixed by trioctahedral micas or vermiculite. Barium ions, with their greater energy 
of hydration, apparently are not readily dehydrated and entrapped by adjacent min- 
eral lattices. Barium also readily rehydrales, forcing the lattices apart when mineral 
surfaces are rewetted after drying. 

Preferential retention of K and NH4 by vermiculite and by weathered mica edges 
is sufficiently dramatic that a sizeable literature has accumulated on this so-called 
fixation reaction. Fixation generally decreases with soil acidification and increases 
with soil liming. This is attributed to the formation of Al and Fe hydroxide interlay- 
ers between mica and vermiculite layer lattices under acid conditions. Such interlay- 
ers prevent the lattices from collapsing completely. Lattice collapse is theoretically 



ELECTROSTATIC CATION RETENTION (CATION EXCHANGE) 215 

necessary to retain fixed cations against exchange by various extracting solutions. 
Fixation is accentuated by drying. 

8.1 .2 Cation Exchange Equations 

To predict the effects of, for example, irrigation, liming, weathering, fertilization, and 
acid rain on soils, it is necessary to predict the exchangeable cation composition in 
equilibrium with this new input. The exchangeable cation chemistry can also provide 
valuable clues about plant elemental deficiencies or imbalances, rates of toxic metal 
movement and attenuation, and tendencies toward soil dispersion. Cation exchange 
equations predict those effects with varying precision. 

Several equations describe cation exchange processes. Each has its own set of 
characteristics and merit. The choice of a particular equation often seems to be sub- 
jective, however, and may be based as much on the investigator's background as on 
any other factor. Lack of familiarity with the units, and of the numerical values of 
exchange coefficients for other equations are major deterrents to adoption of a more 
widespread and uniform approach to cation exchange. 

Certain limitations are inherent in most cation exchange equations: 

1 . Cation and anion exchange are considered separately; acknowledging their si- 
multaneous presence is rare. 

2. The cation or anion exchanger is assumed to possess constant exchange ca- 
pacity. Often, however, the capacity varies with the exchangeable ion, with 
salt concentration, and with pH. 

3. Stoichiometric (1 to 1) ion exchange is generally assumed. Apparent excep- 
tions are usually explained by simultaneous adsorption of molecules or by 
complex ion formation. 

4. Complete reversibility is usually assumed. 

The most general type of cation exchange relationship is a mass action equation: 

CaX + 2Na+ = 2NaX + Ca 2+ (8.7) 

resulting in the reaction coefficient 



(NaX) 2 (Ca 2+ ) 
(CaX)(Na + ) 2 



k = .i. -, (8-8) 



where X denotes the exchangeable form of the cation and parentheses denote activi- 
ties of soluble or exchangeable cations. The major problem in all exchange equations 
is evaluating the activity of the exchangeable cations, since their activities cannot be 
measured or calculated precisely. Equation 8.8 can be rearranged to 

(CaX) - *(Ca 2 +) (8 " 9) 



21 6 CATION RETENTION (EXCHANGE) IN SOILS 

This has been termed a Kerr-type exchange equation. Kerr used ion concentrations 
in place of ion activities, thus tacitly assuming concentrations and activities to be 
directly proportional. Nevertheless, the equation often holds fairly well over narrow 
concentration ranges. The activity coefficient of the divalent cation is more concen- 
tration dependent than that of the monovalent cation, hut the monovalent cation ac- 
tivity coefficient is squared, offsetting much of this variation. The ion activity ratio in 
the soil solution, therefore, may be roughly proportional to the concentration ratios 
in the soil solution over an appreciable range. 

Ion activities in aqueous solution can be estimated from Debye-Hueckel theory 
(Eq. 3. 1 6) or approximated by measurements with specific-ion electrodes. The more 
difficult problem of estimating the activities of adsorbed cations is still unanswered. 
Different assumptions for estimating the activities of exchangeable ions have resulted 
in the several cation exchange equations that are commonly used for exchange be- 
tween ions of different valence. All of these exchange equations reduce to the Kerr 
equation (8.9) when ions are the same valence. The goal of each equation is to pro- 
vide a relatively uniform exchange "constant" (more correctly, exchange coefficient) 
over a wide range of exchangeable cation compositions. The difficulty of this, even 
for ions of the same valence, is apparent from the Kerr equation "constants" plotted 
in Fig. S.2. The exchange coefficient is approximately constant only over a limited 
concentration range for Na-K exchange and over even more restricted ranges for 
Na-Rb exchange. Such variability limits the practical usefulness of most exchange 
equations to relatively small ranges. 

The Gopon equation, proposed in 1933, has found considerable use: 



[NaXj 
[Ca, /2 X] 



k 



[Naj 



rcai 



1/2 



(8.10) 



where exchangeable-cation concentrations are in mmoles of charge per gram (or 
kilogram), and soluble-cation concentrations are in millimoles (or moles) per liter. 

The Gapon equation also uses concentrations rather than activities for the soluble 
ions, and writes the mass action equation with chemically equivalent quantities both 



>8 
15 
12 

G 
3 







' 1 ' 

No-CSy 


1 ' L, ' 1 


_ 


y£ 




_ 


/ 


Na-Rb ^ ,. 


_ 


n 




- 



20 40 60 

% No on the cloy 



100 



FIGURE 8.2. Selectivity coefficient K^ versus exchangeable ion composition for an attapulgite 
clay. (From C. E. Marshall and G. Garcia. 1959. J. Phys. Chem. 63:1663.) 



ELECTROSTATIC CATION RETENTION (CATION EXCHANGE) 21 7 

for colloid exchange sites and exchanging cations. The Gapon equation corresponds 
to the following chemical reaction: 

(Ca), /2 X + Na + =NaX+±Ca 2+ (8.11) 

It differs from the square root of the Ken - equation by including the term [Cai/2X] 
rather than [CaX] 1 / 2 . As was discussed for Eq. 8.9, the successful use of concen- 
tration ratios in solution, instead of activity ratios, is fortuitous over fairly narrow 
(though important) ranges of soluble-ion composition. One example, Na-Ca ex- 
change, is important in irrigated regions. Dispersion and physical deterioration of 
many soils occur if exchangeable Na becomes too high. The Gapon equation is 
unsatisfactory if applied over the entire range of Na-Ca compositions, but works 
fairly well over the range of compositions of most interest to irrigated agriculture. 
The Gapon exchange coefficient is fairly uniform from to 40% exchangeable 
sodium for many irrigated soils of the western United States, at k G = 0.010 to 0.015 
(Lmmor') l/2 - 

Other cation-exchange equations are discussed in Appendix 8.2. Most of these 
have a better theoretical basis than the Gapon equation, but few are as simple to apply 
or visualize. Many workers are willing to sacrifice a little theoretical rigor to gain 
some simplicity in a cation-exchange equation. The Gapon and the Kerr equations are 
the simplest the ion-exchange equations. Gapon adequately predicts cation-exchange 
behavior over practical ranges for many soil systems. 

8.1.3 Diffuse Double Layer 

In air-dry soils, the exchangeable ions can be considered to reside directly on the 
surface of the colloid (Fig. 8.3). The negative charges of soil clays and the layer 
of exchangeable cations make up two slightly separated layers, called a Helmholtz 
double layer. When water is present, however, the cations are no longer so tightly 
held on the surface. The electrostatic attraction of cations is counteracted somewhat 
by diffusion into the aqueous solution. Diffusion tries to equalize the concentration 
throughout the aqueous phase. Figure 8.4 shows the net result of electrostatic attrac- 
tion versus diffusion of cations at two bulk solution salt concentrations, ignoring any 
anion affects. The cation concentration decreases with distance from the negatively 
charged surface. The colloid's negative charge is neutralized by a swarm of positive 
ions in the aqueous phase, the diffuse double layer (DDL). 

Increasing the salt concentration from C\ to Cj reduces the tendency for diffu- 
sion away from the surface and thus shrinks the DDL. The thickness of the DDL is 
loosely defined as the distance over which the solution concentration is affected by 
the colloid's charge. The solution outside the DDL is termed the bulk solution. 

Anion repulsion within the DDL also neutralizes the colloid's negative charge 
by increasing the net positive charge within the DDL. Figure 8.5 shows how anions 
are repelled by the colloid's charge. Assuming for the moment that cations do not 
affect the anion distribution, the anion concentration outside the DDL, C3, is then 
higher than if no repulsion occurred, Co. Since ion charges in the bulk solution must 



21 8 CATION RETENTION (EXCHANGE) IN SOILS 





i 

d 



Helmhollz Loyer 
of Adsorbed 
cotions 



Bulk 
Solution 



DISTANCE FROM COLLOID SURFACE 

FIGURE 8.3. Distribution of monovalent cations and anions near the surface of a typical mont- 
morillonite particle according to the Helmholtz model. (Adapted from D. R. Neilsen et al. 1972. 
Soil Water, p. 45, by permission of the American Society of Agronomy and Soil Science Society 
of America.) 



< 



Ld 

u 

Z 
O 
<J 

o 
o 





DISTANCE FROM COLLOIDAL SURFACE 

FIGURE 8.4. The distribution of cations away from a negatively charged soil surface at two 
cation concentrations, with effects of anions disregarded. The cation exchange capacity is pro- 
portional to the area between the curves and their corresponding dashed lines. 



ELECTROSTATIC CATION RETENTION (CATION EXCHANGE) 



219 




DISTANCE FROM COLLOIDAL SURFACE 



FIGURE 8.5. Distribution of anions near a negatively charged surface at two anion concentra- 
tions, C 4 and C5, disregarding cation effects. 



balance, this means that anion repulsion increases the salt concentration in the bulk 
solution. Increasing the anion concentration to C4 shrinks the DDL, as it did for 
cation attraction, though in this case the effect arises from increased anion diffusion 
toward the surface. The effect of anion repulsion is mostly interesting in the labo- 
ratory, because the contribution of anion repulsion to soil behavior is small except 
when bulk solution salt concentrations are > 1 M, where few organisms can survive. 

Figure 8.6 shows the combined result of cation attraction, anion repulsion, and ion 
diffusion on the cation and anion distribution next to a negatively charged particle. 
The solution near the surface has an excess of cations and a deficit of anions. The 
thickness of the DDL decreases with increasing cation or anion charge. The effect 
of anion charge is less significant, because fewer anions are in the DDL. The DDL 
shrinks with increasing cation charge because fewer ions are necessary for charge 
neutralization and the more highly charged cations are attracted more strongly to the 
colloid. Diffusion, on the other hand, results from ion concentrations rather than ion 
charge. 

Table 8.4 gives values for an arbitrary double-layer "thickness" for monovalent 
and divalent cations at three bulk solution salt concentrations. Double-layer thickness 
varies inversely with the square root of the bulk solution salt concentration or the 
valence of the exchangeable cation. These thicknesses are small compared to the 
diameters of soil pores, which are on the order of 1000 to 50 000 nm, but they are of 
the same magnitude as water film thicknesses in relatively dry soils. 

If the DDL contained only those cations necessary to neutralize the colloid charge, 
the anion concentration would be zero within the DDL. Because diffusion contin- 
ually drives anions toward the colloid surface, however, the total negative charge 
within the DDL is that of the anions plus the colloid's charge. Cations within the 
DDL must neutralize both sources of negative charge. The cations that neutralize 



220 CATION RETENTION (EXCHANGE) IN SOILS 





3.0 




2.0 




1.0 




0.3 




02 


5 


l 


o 


0.05 


i- 




< 




1- 


002 


S 




UJ 


0.01 






? 




O 





0001 



00001 



- 


? i 




1 






- 


A 


^VC AT IONS 










- 


_ 


\^ANI0NS 


Co 


= 1 






_ 


-/ 






__CATI0NS 


Co 


= 0.01 


- 


- 


/^ANIONS 

1 . 1 


I 






- 



2 4 6 8 10 12 

DISTANCE (nm) FROM COLLOID SURFACE 



FIGURE 8.6. Distribution of monovalent cations and anions near the surface of a montmoril- 
lonite particle. (Adapted from D. R. Nielsen et al. 1 972. So/7 Water, p. 45, by permission of the 
American Society of Agronomy and the Soil Science Society of America.) 



the colloid charge are "exchangeable"; the remainder are "soluble," because they 
neutralize the anions that have diffused into the DDL. The total positive charge in 
the DDL must exactly equal the total negative charge of that region. For positively 
charged colloids, the behavior of cations and anions in the DDL and bulk solution 
are reversed. Anions are attracted to the surfaces and cations are repelled. 



Table 8.4. Approximate "thickness" of a typical soil colloid double layer as a function of 
electrolyte concentration' 3 



Bulk-Solution 
Concentration of 
Cations (mol charge L~ ' ) 



"Thickness" of the Double Layer (nm) for: 



Monovalent 
Cations 



Divalent 
Cations 



10" 

10 

10 



-3 

-1 



10 
I 
0. 



5 

0.5 

0.005 



"Adapted from H. Van Olphcn. 1963. An Introducton to Clay Colloid Chemistry. Inlerscience, New York. 
Reprinted by permission of John Wiley & Sons, Inc. 



STRONGLY-RETAINED CATIONS 221 

8.2 STRONGLY-RETAINED CATIONS 

The strongly retained cations in soils include many of the essential microelements 
and also the "toxic" cations. The concentrations of these ions in the soil solution are 
low and they are apparently retained by two means. One group is the cations that in 
aqueous solutions precipitate as insoluble oxides and hydroxyoxides. The root zone 
of a typical agricultural soil might contain as much as 300 000 kg ha - ' of Fe and Al, 
but their plant availability is only a few kg ha~' . 

The second and smaller group is those cations that also are insoluble in pure aque- 
ous solutions, but that tend to associate in soils with soil organic matter and sulfide. 
These cations are the weak and intermediate Lewis acids of Table 3-7 — Cu'~ 2+ , 
Cd 2+ , Hg l_2+ , Ni 2+ , Zn 2+ , and Fe 2+ . Soil retains these cations by hydroxyoxide 
precipitation/adsorption and by soil organic matter adsorption in varying degrees. 
The amounts of the weak Lewis acids in soils are small; the root zone contains on 
average about 300 kg ha -1 of Pb and < 1 kg ha -1 of Cd and Hg but the variation is 
large. The amounts of these ions absorbed by plants is a tiny fraction of the total. 

Soil retention of strongly retained ions generally increases with pH. Above pH 7, 
the effect of increasing pH on ion movement, plant availability, and chemical ex- 
tractability lessens. Molybdenum is an exception: MoO 2- reacts strongly with Ca 2+ 
and precipitates at higher pHs. If these ions are added to soils, as in municipal and 
industrial wastes, contaminated water, fly ash, and so on, most tests have shown 
that the amounts retained and the strength of that retention increases with time. The 
amounts retained and the strength of retention increase rapidly at first; the rate then 
slows over periods ranging from days to months. 

The generalization that heavy metal retention increases with time may be wrong 
for weak Lewis acids applied in organic wastes to soils. One recent experiment with 
Mo, Zn, and Cd added to soils in sewage sludge showed that their plant availability 
remained unchanged for many years after application. The plant uptake of these ions 
was linearly proportional to soil content even after 23 years since the last addition. 
The amount of Zn available was 0.5 to 3% of that added initially. The amount of 
available Cd was 4 to 18% of that added. The reasons for the long-term and high 
availability in this case is uncertain. 

8.2.1 Oxide-Retained Cations 

The relative retention of divalent ions by amorphous Fe hydroxyoxides is 

Pb > Cu > Zn > Ni > Cd > Co > Sr > Mg 
Retention of the divalent cations by Al hydroxide is slightly different; 

Cu > Pb > Zn > Ni > Co > Cd > Mg > Sr 
Retention by silica is 

Pb > Cu > Co > Zn > Ni = Cd > Sr > Mg 



222 



CATION RETENTION (EXCHANGE) IN SOILS 



Although differing in detail, the retention by these three major soil components is 
rather similar. The measurements were done in the absence of organic matter or 
sulfide ions, so only the attraction to O 2- -dominated surfaces was determined. 

The retention by silicate and hydroxyoxide minerals and surfaces is by O 2- ions. 
Since the oxide ion interacts strongly with H + , the cation retention is strongly pH 
dependent. Figure 8.7 illustrates both the pH and time dependence of the retention 
of Cd 2l ~ by Fe hydroxyoxide. At pH < 6, the amount of Co 2+ retained by Fe hy- 
droxyoxide increased with pH and time. At pH > 6.5, Co~ + retention was complete 
within 2 weeks. 

The initial retention shown by the black triangles is low as acidity decreased to 
pH 7, whereupon retention increased sharply, the so-called adsorption edge. When 
the solid was treated later with acid, the amount of Co 2+ released back to the solu- 
tion was less than the amount added, and the release decreased as time of aging the 
Co-Fe hydroxyoxide mixture increased. Other ions react similarly with silicates and 
hydroxyoxides. The pH of the adsorption edge and the effect of time differ some- 
what, but the behavior is rather general. Plant deficiencies of these essential mi- 
croelements, such as Fe chlorosis and Zn deficiency, may occur above soil pH 8. The 
deficiencies have been noted in fruit and nut trees and in some varieties of sorghum. 
The deficiencies are in arid regions under irrigation in which the faster growth puts 
a greater stress on soil availability. Native plants growing under native conditions do 
not show deficiency symptoms. Plant deficiencies are more complicated than simple 
soil unavailability, but the deficiencies can usually be overcome by acidifying the 
soil. 

Conversely, when the concentrations of these cations are too high — Al toxicity in 
acid soils or Fe and Mn toxicity in rice paddies — raising the soil pH by liming is ef- 



100 



BO- 



'S 60- 

0) 

o 
O 



20- 







** 


*~~** 1 


- 




A CTJ 

Jfe 




□ 


Q 


■4 


Adsorption pH edge 


© 

• 

I A 


4 iii 

^iii 


A aging time = wk 

Desorption 

9 aging time = 2 wk 

O aging time = 9 wk 

□ aging time = 16wk 

i i i i 



PH 



10 



12 



FIGURE 8.7. Fractional adsorption of Co 2+ to hydrous Fe-oxide (HFO) as a function of pH 
and HFO-Co 2+ aging time. (From C. C. Ainsworth, J. L. Pilou, P. L. Gassman, and W. G. van- 
der Aluys. 1994. Soil Sci. Soc. Am. J. 58:1615.) 



STRONGLY-RETAINED CATIONS 223 

fective. Since 2 ~ ligands are so prevalent in soils, they can retain/adsorb/precip'rtate 
large amounts of these ions if pH > 6. Changing the soil pH or foliar spraying to 
circumvent the soils is generally more effective than adding these microelements as 
soil fertilizers. That also means that soil pollution by these ions is readily overcome 
by increasing the soil pH. The ions are immobile and unavailable for plant or mi- 
crobial uptake. Liming contaminated soils, covering them with clean soil, or mixing 
them with clean soil underneath, is often sufficient to overcome any hazards. Regu- 
latory agencies realize that the mere presence of an ion in soils does not necessarily 
constitute a hazard to humans. Soils have a high capacity to reduce high additions of 
these elements to their plant availability in native soils. 

The retention of this group increases, that is, mobility and plant availability de- 
creases, rapidly at first and the rate slows with time. This behavior is similar to dif- 
fusion and suggests that the mechanism is the slow transfer of surface ions into the 
weathered layer on soil particles. The initial rate is rapid because the surface concen- 
tration is relatively high and the diffusion path length is short. As diffusion inward 
proceeds, the surface concentration decreases and the diffusion path length increases. 
The shallower concentration gradient slows the rate of further cation diffusion. 

Although the cations retained primarily by aluminosilicates and hydroxyoxides 
in soils, that is, the hard Lewis acids, are controlled by interaction with O 2 ' and 
OH"" ligands on soil particle surfaces, the aqueous solubility of these cations is usu- 
ally much less than that predicted by the solubility products of their pure hydrox- 
yoxides. The big differences between solubility products and ion activity products 
(IAP) indicate that soils retain the cations more strongly than their own pure hy- 
droxyoxides. How soils can accomplish this is still uncertain. Although the cations 
(it better into their own hydroxyoxide structures than into aluminosilicate or major 
Al, Fe, Mn, and Ti hydroxyoxide structures, soils nonetheless retain the ions very 
strongly. 

One school of thought maintains that certain "sites" on soil surfaces can retain 
these cations strongly; radiographs show the cations are both bunched and spread 
out on soil surfaces. Another school suggests that these adsoiption sites are where the 
cations can mix as solid solutions with the other ions on the surfaces. The free energy 
of mixing on the surfaces (Appendix 3.2) is responsible for the strong retention rather 
than any uniquely favorable adsorption spots on soil particle surfaces. In any case, 
soil retention can reduce the aqueous solubility of these ions to well below that in 
equilibrium with their pure hydroxyoxides. 

The models that predict soil solution concentrations by their oxide/hydroxyoxide 
or other solubility products (K sp ) have usually been only qualitatively successful. 
Predictions of aquated Al and Fe concentrations have sometimes been successful, 
probably because their hydroxyoxides were a major component of the solid phase 
and therefore unaffected by the effects of solid solution mixing. For trace metals, the 
models have been less successful. In some cases the measured IAP has been similar 
to the K sp of a pure solid that has not yet been identified in soils. Pb solubility in 
the soil solution has been linked to the solubility of Pb phosphate, for example. This 
must still be tested by measuring the Pb response to added phosphate to see if the 
agreement of IAP and K sp is coincidental rather than causal. 



224 CATION RETENTION (EXCHANGE) IN SOILS 

Another approach is simply to statistically relate the ion concentration in the soil 
solution to the total amount in the soil. In 100 British soils, the Pb concentration in 
the extracted soil solution closely followed the following equation: 

l«g Pbtou.1 - log Pbsoiisoiinion = 1.30 + 0.55pH (8. 12) 

In 30 British soils, the soil Cd concentrations followed 

log Cdtotai - 1 .09Cd soilsolution =1.11+ 0.38pH (8. 1 3) 

Such relations depend on the analytical procedure employed. The relations do not 
support one retention mechanism over another but are useful for regulatory purposes. 
The soil solution concentrations ranged from 3.6 to 3600 fxg Pb L _l and 2.7 to 1280 
/ig Cd L -1 . The amounts in the extracted solutions were less than 1% of the total 
amounts in the soils, indicating how effective soils are in reducing availability of 
these toxic ions. Although several retention mechanisms may be involved for Pb 
and Cd, all the mechanisms apparently respond to pH. Neither equation supports the 
generalization of increasing retention with time. 

8.2.2 Cations Retained by Soil Organic Matter 

The soft Lewis acids (Section 3.5) — Cd, Cu, Zn, Hg, Pb 2+ , and to a lesser extent 
Fe 2+ and Mn 2+ — tend to be more affected by soil organic matter and sulfide. Some 
generalities about these cations are the following: 

1 . Soil retention is less pH sensitive than for hard Lewis acids. 

2. Retention may be less time dependent. 

3. Fraction retained, and strength of retention, decreases as concentration in- 
creases. 

4. The ion activity products of the cation's hydroxyoxide are less than their hy- 
droxyoxide solubility products. 

Although soft Lewis bases are associated with reducing (anaerobic, low oxygen) 
conditions, normal (aerobic) soils contain sufficient amounts of organic matter to 
retain the low amounts of these trace metal ions in soils. 

The soft Lewis acids and bases are among the very toxic ions. The soft Lewis 
acid-base situation that has created considerable concern is in municipal and indus- 
trial landfills. The worry is that the strong reducing conditions will create soluble soft 
Lewis acid-base ions, complex ions and molecules that will leach out of the landfills 
and into groundwater. To counteract this problem, clay and plastic liners are being 
installed beneath landfills and water-impermeable caps are being placed above the 
landfills to entomb the wastes. The slow migration of ions beneath natural swamps, 
another strongly reducing condition, suggests that the severity of the problem may 
be exaggerated. The problem, however, has made people more aware of proper and 



DIFFUSE DOUBLE-LAYER THEORY 225 

improper use of soil as a disposal/recycling medium. The idea of entombing our 
wastes in "secure" landfills seems naive; wise treatment would seem to be a better 
alternative. 



APPENDIX 8.1 DIFFUSE DOUBLE-LAYER THEORY 

The Guoy-Chapman theory, derived concurrently by them in the early 1900s, is the 
basis of describing the DDL on charged colloid surfaces. Their assumptions are sim- 
ilar to those used later and more successfully by Debye and Hueckel to describe ion 
activities in the much simpler case of aqueous solutions. Gouy-Chapman theory as- 
sumes that ( 1 ) exchangeable cations exist as point charges, (2) colloid surfaces are 
planar and essentially infinite in extent, and (3) surface charge is distributed uni- 
formly over the entire colloid suiface. These assumptions inaccurately describe ac- 
tual systems, but the theory of the DDL works surprisingly well for soil colloids. 
Apparently, many of the errors inherent in the assumptions tend to cancel each other. 
Cations are attracted toward, and anions are repelled from, negatively charged soil 
colloids. Such interactions follow Coulomb's law: 

qq'K 

where F is the force of attraction or repulsion (newtons), q and q' are the elec- 
trical charges (coulombs), K is a proportionality constant {— 8.9 x 10 9 for these SI 
units), r is the distance of charge separation (meters), and D is the dielectric constant 
(= 78 for water at 25° C). The strength of ion retention or repulsion increases with 
increasing ion charge, with increasing colloid charge, and with decreasing distance 
between the colloid surface and either the source of charge or the soluble ion. 

The increased cation concentration in the DDL develops a countertendency for 
cation diffusion away from the surface. The diffusion tends to equalize cation con- 
centrations throughout the solution phase. Combining the equations for cation attrac- 
tion and diffusion yields the Boltzmann equation: 

|=cxp(-Z^) (8.15) 

where C is the concentration of an ion at a specified distance from the charged sur- 
face, C'o is the concentration of the ion in the bulk solution, Z is the valence of the 
ion, e is the unit of electronic charge, i/r is the electrical potential of the colloid at 
the specified distance, k is the Boltzmann constant (the gas constant per molecule), 
and T is the absolute temperature. Equation 8.15 describes the distributions of both 
cations and anions in the double layer, provided that tJ/ is made negative because of 
the net negative charge of most soil colloids. 

The treatment of double-layer phenomena is straightforward when the change of 
electric potential with distance from the colloid surface can be adequately estimated. 
This distribution can be considered to arise from the termination of individual lines of 



226 CATION RETENTION (EXCHANGE) IN SOILS 

force from the colloid when they encounter cations in the double layer. Solutions of 
the equations describing electric potential distribution are, however, mathematically 
complex. 

The DDL is generally treated quantitatively in either of two ways. The more stan- 
dard approach is to regard soil colloids as having constant surface charge but variable 
surface potential (constant charge colloids). The distribution of potential thus varies 
with bulk solution salt concentration and with average valence of the counter (ex- 
changeable) ions, but the charge (CEC) of the soil remains constant. 

A second approach is to treat the colloid as having constant surface potential but 
variable surface charge (constant potential co/loids). This behavior is common for 
colloids such as gold sols or glass surfaces and for soils having predominantly pH- 
dependenl charge. In most soils, however, the main potential-determining ions are 
H and OH ions. Hence, the charge remains virtually constant as long as the pH is 
held constant, unless the salt concentration or the exchangeable cation composition 
changes markedly. Unlike for variable potential colloids, the charge of constant po- 
tential colloids varies appreciably and predictably with the salt concentration of the 
bulk solution. Adsorption of certain anions can also change the colloid charge. 

Hydroxyoxide surfaces often appear to have concentration-dependent anion ex- 
change capacities, but such behavior can also be explained by the collapse of the 
double layer at high salt concentrations. Under these conditions, positively charged 
sites are no longer masked by the DDLs of the predominantly negatively charged soil 
matrix. 

Fully expanded double layers are rare in field soils. Double-layer expansion nor- 
mally is restricted to thin water films on colloid surfaces or by interactions with 
double layers on adjacent soil particles within aggregates. Figure 8.8 represents such 




FIGURE 8.8. Electrical potential and ion concentration between interacting negatively charged 
platelets. 



DIFFUSE DOUBLE-LAYER THEORY 227 

a restricted double layer. The distributions of cations, anions, and electrical potential 
in Fig. 8.8 are assumed to be symmetrical between two vertical colloid particles. The 
cation concentration decreases to C^\ and the anion concentration increases to C^~, 
the solute concentrations at the midplane between adjacent particles. Similarly, the 
electrical potential varies from A s , the surface potential of the colloid, to Vt» m e 
midplane electric potential. The excess of cations at the midplane compared to the 
bulk solution causes an osmotic gradient. This, in turn, causes water imbibition, or 
swelling, of the colloid. Water imbibition continues until the tendency to swell is 
balanced by interparticle bonds; that is, until the osmotic potential at the midplane 
equals that in the bulk solution, or until swelling is retarded by the lack of additional 
water. For most soils, obvious swelling of the entire matrix is uncommon, but it can 
be pronounced for highly montmorillonitic soils, such as Vertisols. 

In theory, one can calculate the distribution of electric potential within the double 
layer for any combination of colloid charge, salt concentration, counter-ion valence 
and interparticle distance. The Boltzmann equation (8.15) can then be used to calcu- 
late cation and anion distributions. From such distributions, cation exchange, colloid 
swelling, and anion repulsion can be inferred but the calculations are complex, te- 
dious, and often only approximate. 

In practice, a set of curves developed by Kemper and Quirk (Fig. 8.10), yields 
approximate electric potentials as a function of distance from the colloid surface. 
Such potentials can then be substituted directly into the Boltzmann equation to infer 
cation and anion distributions. Yi, is the scaled electric potential (equal to -Zei///kT 
of Eq. 8.15) at the midplane between interacting colloids, V is the surface charge 
density in coulombs m~ 2 (96.5 times the ratio of CEC, in mmoles charge kg -1 , 
divided by the specific surface, in m 2 kg" ' ), Z is the valence of the exchangeable 
cation, Cq is the molar salt concentration in the bulk solution, and x is the distance (in 
nm) from the midplane between colloids to the plane at which the ion concentration 
is to be calculated. 

Figure 8.9 is designed for partially expanding (interacting) double layers. A value 
of Yb — 0.01, however, is normally a satisfactory approximation for noninteracting 
colloid surfaces. To estimate the thickness of the water film, move across the fig- 
ure from right to left on the line representing r s /(Co)^ 2 axis to obtain the distance 
from midplane to surface (x). When the distance to the midplane is known and Yb 
is to be estimated, on the other hand, move from right to left across the figure on a 
FbACo) 1 ^ 2 line to the appropriate value of Zx(Co)^ 2 . Then estimate Xbby interpo- 
lating between Y\, lines. 

In either of the above cases, the intersection of horizontal and vertical lines, when 
projected horizontally onto the Y axis, determines the surface potential Y s . Values of 
the electric potential for the system can vary only between Y\, and y s . The Y/, line 
then yields values of Y corresponding to selected x values. The cation and anion 
concentrations can then be calculated as functions of distance from the colloid sur- 
face or from the midplane between interacting colloids by using these Y values and 
Eq. 8.15. When estimating anion distributions from the Boltzmann equation, Y must 
first be multiplied by Z(_)/Z( + ) to obtain values that decrease appropriately with 
proximity to the colloid surface. 



228 CATION RETENTION (EXCHANGE) IN SOILS 

Y b (vertical curved lines) 



0.04 002 0.01 

r 




t 



FIGURE 8.9. Scaled potential (/(, = Zef/kT) as a function of solution concentration and of 
distance from the midplane, charge density, and soil colloid surfaces. (From W. D. Kemper and 
W. P. Quirk. 1970. Soil Sci. Soc. Am. Proc. 34:347-350.) 



To derive Fig. 8.9, Kemper and Quirk assumed symmetric electrolytes (e.g., NaCl 
and CaSC>4, but not CaCh). Such an assumption is unreasonable for many soil so- 
lutions, but causes relatively minor errors in most cases. The effects of symmetric 
versus nonsymmetric electrolytes on electrical potential distribution are consider- 
ably less than the effects of ion valence, total salt concentration, or surface charge 
density. 

Stern improved the Gouy-Chapman theory of the DDL by assuming that some 
ions are lightly retained immediately next to colloid surfaces in a layer of specifi- 
cally adsorbed or Stern- layer cations. The double layer is diffuse beyond this layer. 
A satisfactory approximation of the Stern model can be made by assuming that the 
specifically adsorbed ions quantitatively reduce the surface density of the colloid. 
The diffuse portion of the double layer then is assumed to develop on a colloid sur- 
face of correspondingly reduced charge density. Sample Stern-modification calcula- 
tions for a series of monovalent cations are shown in Fig. 8.10. Relatively few of the 



CATION EXCHANGE EQUATIONS 229 




0.5 1.0 1.5 2.0 0.5 1.0 1.5 2.0 0.5 1.0 1.5 2.0 

DISTANCE FROM PLATELET SURFACE (nm) 

FIGURE 8.10. Calculated cation distributions near a mineral surface. (From I. Shainberg and 
W. D. Kemper. 19S6. Soi/ Sci. Soc. Am. Proc. 30:707-713.) 



strongly hydrated lithium ions are strongly adsorbed in the Stern layer. Most lithium 
ions are in the diffuse layer instead. The opposite trend is evident for the weakly 
hydrated potassium ions. Shainberg and Kemper treated the implications and appli- 
cations of the model, and consequences of its assumptions, during the mid 1960s. 



APPENDIX 8.2 CATION EXCHANGE EQUATIONS 

As indicated in Section 8.4, the primary difference between various cation exchange 
equations is their differing treatment of the activities of exchangeable cations. 
Vanselow, for example, assumed that the activities of exchangeable cations were 
proportional to their mole fractions. This is equivalent to saying that ions on soil 
colloid surfaces behave as if in ideal solution (Appendix 3.2). The mole fraction of 
an ion in a binary system is 



mole fraction of species a 



«a + "b 



(8.16) 



where n is the number of moles per unit volume or per unit mass. Substituting this 
assumption into a Kerr-type expression (Eq. 8.8), yields the Vanseiow-Argersinger 
equation for monovalent-divalent cation exchange 



[NaX] 2 



(Na"h 2 



[CaX] [NaX + CaX] v (Ca 2+ ) 



(8.17) 



Here brackets indicate exchangeable ion concentrations (mmol kg ' ) and parenthe- 
ses denote soluble ion activities (mmol L _1 ). Since the left-hand side of the equation 



230 CATION RETENTION (EXCHANGE) IN SOILS 

is dimensionless, K v has units of L mmol -1 . The Vanselow-Argersinger equation 
has been used extensively to characterize cation exchange on simple, relatively uni- 
form exchangers. In the surface chemistry literature, this equation has been used 
to calculate so-called thermodynamic exchange constants. However, these constants 
generally are simply averaged values of Vanselow coefficients over a range of ex- 
changeable and soluble ion compositions. They rarely describe the ion distribution 
at any specific composition precisely. 

Davis developed an equation similar to the Vanselow equation from statistical 
thermodynamics. Electrostatic forces between colloid surfaces and adsorbed cations 
were calculated for various surface configurations of charge sites. These sites were 
assumed to be neutralized by individual adsorbed ions. Hence, the model resembles 
most closely the Helmholtz model of the double layer with the charge of cations on 
the surface assumed to be just equal to the number of colloid charges. The resultant 
equation is 

^ K d ^- (8.18) 



[Ca][NaX + 9C aCaX] (Ca 2+ ) 

where 

qc* = Z CA --^ + - (8.19) 

Here Y is the number of nearest-neighbor (closest) charge sites, and Z the cation 
valence. The main difference between the Davis equation (8. 1 8) and the Vanselow 
equation (8.16) is the specific ion factor q\ for the divalent cation. For monovalent 
ions, q\ is unity. 

Despite the entirely different theoretical bases of the Davis and Vanselow equa- 
tions, each produces essentially the same expression for ion exchange. Values for 
the specific ion factor (q-,) include 1.0 for a linear array of cation exchange sites 
(two nearest-neighbor exchange sites), and 1 .67 for a close-packed array (six nearest- 
neighbor sites) (Fig. 8.11). The linear array gives results that are numerically equal to 
the Vanselow equation. Krishnamoorthy and Overstreet tested several configurations 
and concluded that a large number of soils behaved as if the exchange sites were in 



Linear Open-packed Close-packed 

qj = I.O qi = l.5 qj =1-67' 

(a) (b) (c) 

FIGURE 8.11. Surface arrays of cation exchange sites used in the Davis (statistical thermody- 
namics) equation for cation exchange. 



CATION EXCHANGE EQUATIONS 231 

open-packed array (q\ = 1.5). Because Davis delayed publishing his work, Eq. 8.18 
has sometimes been called the Krishnamoorthy-Overstreet equation instead. 

Eriksson and Bolt used another approach to develop the Eriksson, or double layer, 
exchange equation: 

T = FT* sinh , / m- < 8 - 2 °) 

r TVp r + 4vWM 2 

where T is the colloid charge density (mmol(— ) m -2 ); V\ is the portion of the charge 
neutralized by the monovalent cation; r is the reduced ratio (= M\ l(M%) 1/ ' 2 ), where 
M\ and Mi are the activities of monovalent and divalent cations in the bulk solu- 
tion; /Sis a constant equal to 1.08 x 10 20 nimol -1 for aqueous systems at 25° C; 
and Dd = cosh Kb ( tne hyperbolic cosine of exp — Ze\jr/kT) from Eq. 8.15 at the 
midplane between adjacent colloid particles). In practice, v& commonly is set equal 
to unity. This is equivalent to no interaction between adjacent particles or to infinite 
interparticle distance (realistically, to a particle separation of a few tens of nanome- 
ters). 

An advantage of the double-layer equation is that it calculates ion distributions 
entirely from physically measurable parameters, such as CEC, surface area, and bulk 
solution solute concentrations. Disadvantages are that these measurements are rather 
involved, and that the inverse hyperbolic sine makes it difficult to visualize the ef- 
fects and implications of changes on cation exchange. Generally, experimental sur- 
face charge densities for soils and clays must be multiplied by a factor of 1.2 to 1 .4 
to make experimental results and this equation's predictions agree satisfactorily. The 
need for this modification may result either from errors in the surface area estimates 
or because different portions of the surface affect the exchange capacity and surface 
area measurements differently. Nonetheless, Eq. 8.20 permits a wide variety of cal- 
culations almost entirely from measurable soil properties. The Kerr, Vanselow, and 
Davis equations, in contrast, require measurement of empirical exchange coefficients 
for different sets of experimental conditions to make such predictions. 

Workers at the U.S. Salinity Laboratory substituted the sum of Ca plus Mg for 
the exchangeable and bulk-solution Ca concentrations in the Gapon equation (8.10). 
This yields 

LNaXl , [Na+] 

= * J. „,x,w, (8.21) 



[CaX + MgXJ [Ca 2+ + Mg 2+ l'/2 

The fCa 2+ + Mg 2+ 1 term was necessary because many early water analyses did not 
distinguish between the two ions. The left side of Eq. 8.21 was termed the ESR 
(exchangeable sodium ratio). The solution concentration ratio on the right side was 
termed the SAR (sodium adsorption ratio). The SAR is written as [Na + ]/([Ca 2+ +■ 
Mg 2+ ]/2)'/ 2 when the concentration units are millimoles of charge per liter. The 
reduced ratio (/) of the double-layer exchange equation (8.19) is equal to the SAR 
divided by (1 000) l/2 . 

From analyses of saturation extracts and exchangeable cation concentrations for a 
large number of soils from the western United States, the statistical relation of ESR 



232 CATION RETENTION (EXCHANGE) IN SOILS 

and SAR was found to be 

ESR = -0.01 + 0.015 (SAR) (8.22) 

This is equivalent to the Gapon equation, with an exchange constant of 0.015, except 
for the small negative intercept. For many applications, the intercept is negligible. 
Soils outside the principal irrigated portions of the western United States, such as 
the irrigated tropics, may have Gapon constants appreciably different from 0.015. 

Another early approach to ion exchange was that of the Donnan equilibrium. This 
concept described a system in which a solution and suspension were separated by a 
membrane that is permeable to ions but impermeable to the exchanger or clay. An ex- 
ample is filter paper separating a soil suspension and its extract. In a "micro-Donnan" 
system, each soil colloid particle with its ion swarm is regarded as being separated 
from the bulk solution by an imaginary membrane. Basic Donnan equilibria apply to 
homovalent exchange in soils and can also be used to explain dilution effects during 
exchange between ions of different valence. 

Eriksson applied Donnan equilibrium calculations to heterovalent exchange, rea- 
soning that clays in a salt solution could be thought of as an ion species restricted 
from free diffusion. His equation was 

<^ = ^ (8-23) 

(A)} (A)£ 

where i and o refer to ions inside the clay phase and outside (in the bulk solution). 
This is equivalent to a Kerr equation with K K = I . Concentrations were multiplied 
by activity coefficients for the solution phase, and calculated from the amount of ad- 
Table 8.5. A comparison of exchange coefficients for several cation exchange 
equations, as calculated from the ammonium-calcium exchange data of Table 8.2 





1 


Exchange Coefficient at an Ammonium 








Concentration of: 




Equation 


0.04 M 




0.01 M 


0.005 M 


0.0025 M 


Kert (using concentrations) 


1.91 x 10" 


-3 


1.81 x KT 3 


1.58 x 1(T 3 


2.30 x 10 -3 


Kerr (using solution activities) 


1.33 x 10" 


-3 


1.51 x 10 -3 


1.37 x 10~ 3 


2.07 x I0- 3 


Gapon (using concentrations) 


1.77 




1.07 


0.97 


1.17 


Vanselow 


2.22 




2.50 


2.35 


3.60 


Davis (<fca = 1 -5) 


1.75 




1.88 


1.73 


2.63 


Davis (0ca= 1.67) 


1.63 




1.73 


1 .59 


2.41 








Exchangeable NH^j" Percentage 




Erikkson 












Predicted" 


26.1 




20.2 


17.9 


14.4 


Measured 


29.8 




20.8 


16.6 


15.2 



"Assuming a surface area of 800 x I0 3 m 2 kg 



DETERMINATION OF CATION EXCHANGE CAPACITY AND EXCHANGEABLE CATIONS 233 

sorbed ions divided by the volume of exchanger for ions in the adsorbed phase. This 
fixed the activity coefficients for the adsorbed-phase ions as well. Today, Donnan 
equilibrium is used only as a first approximation to DDL theory. The main objection 
to Donnan theory is the large error involved in predicting concentrations at varying 
distances from the particle surface with a single ion ratio. Donnan theory predicts 
only an average activity ratio, which may err by a factor of two or more. Other 
criticisms have arisen from its inability to adequately predict the properties of clay 
suspensions. 

Table 8.5 compares different exchange coefficients calculated from the data for 
ammonium-calcium exchange in Table 8-2. The simple Gapon equation (8. 10) yields 
the most uniform exchange coefficient for this set of data; the Eriksson equation's 
predictions also agree well with the measured values. Bond and Verburg (1997) ap- 
plied the various ion equations to the more complicated case of ternary (Ca-K-Na). 
Their slight modifications of the 1918 work by Rothmund and Kornfeld yielded the 
most consistent exchange coefficients in their study. Snyder and Cavallaro (1997) 
applied a single-phase mixture approach to NH^"-Ba 2+ -La 3+ exchange on clays. 



APPENDIX 8.3 DETERMINATION OF CATION EXCHANGE CAPACITY 
AND EXCHANGEABLE CATIONS 

When cation exchange relations are measured, both the total and the relative quan- 
tities of the exchangeable cations are required. To determine the total quantity of 
exchangeable cations, the cation exchange reaction is normally forced toward com- 
pletion by either of two approaches. In one, the soil sample is exhaustively leached 
with a solution that contains a replacement, or index, cation to be placed on all ex- 
change sites. Ions removed by the leaching are then analyzed to determine the initial 
exchangeable cation composition. A second approach involves repeated batch wash- 
ings (several cycles of adding replacement cation, shaking, centrifuging, and decant- 
ing the supernatant solution). Analysis of the combined supernatant solutions yields 
the amounts of exchangeable ions initially present. The large excess of replacing ion 
in each batch-washing step drives the reaction toward completion. 

In determining the CEC, a soil saturated with a single index ion is washed free of 
soluble salts, often with alcohol to keep the soil flocculated and to prevent loss of the 
index cation by hydrolysis- 

NaX + H 2 = HX + Na + + OH - (8.24) 

where X represents the clay and Na + is the index cation. The H + clay is unstable 
and rapidly breaks down: 

HX + Al(OH) 3 = Al(OH) 2 X + H 2 (8.25) 

Hydrolysis would yield a low estimate of the CEC. The index cation is then extracted 
from the soil with still another salt solution and measured to give the CEC. 



234 CATION RETENTION (EXCHANGE) IN SOILS 

The salt used to furnish the index cation should be relatively soluble in the alcohol 
used for sample washing. The low solubility of NaCl in ethanol has been a frequently 
overlooked source of error. The salt then dissolves during the subsequent extraction 
step and yields an anomalously high CEC value. Soils containing large quantities of 
hydroxyoxide or amorphous minerals may also retain salts in particle micropores, so 
that washing does not completely remove the salts. This also yields high CEC values. 

To eliminate the problems associated with the washing step, Okazaki, Smith, and 
Moodie proposed a CEC procedure in which salts are not removed between the in- 
dex cation saturation and extraction steps. Rather, the anion of the salt providing the 
index cation is analyzed in the final extract. In accordance with electrical neutrality, 
the CEC is then equal to the total quantity of index cations removed during extrac- 
tion minus the quantity of index anions removed simultaneously. The main potential 
source of error from this procedure arises from anion repulsion, if the quantity of 
index salt remaining after saturation is merely calculated from the weight of solution 
retained and its initial (or average) concentration. This error minimized if the index 
solution is lowered to approximately 0.1 M during the final two saturation washes. 
The error is eliminated if the quantities of index salt are analytically determined in- 
stead. 

The concentration of the index salt solution should not be high. Early measure- 
ments using 1 M salt solutions, to insure complete replacement and flocculation, 
yielded low CEC measurements because anion repulsion is significant at these con- 
centrations and neutralized a significant portion of the colloid's charge. The CEC 
measurement is one of many examples in soil chemistry of the complexity of a seem- 
ingly simple experiment. 

BIBLIOGRAPHY 

Bond, W. J., and K. Verburg. 1997. Comparison of methods for predicting ternary exchange 
from binary isotherms. Soil Sci. Soc. Am. J. 61:444-454. 

McBride, M. B. 1990. Reactions controlling heavy metal solubility in soils. Adv. Agron. 10:1— 
59. 

Snyder, V. A., and Cavallaro, N. 1997. The thermodynamic theory of ion exchange: a single- 
phase mixture formulation. Soil Sci. Soc. Am. J. 61:36-43. 
Thompson, H. S. 1850. On the absorbent power of soils. J. R. Agr. Soc. 11:6874. 



QUESTIONS AND PROBLEMS 

1. The following distribution of cations and anions exists near a soil colloid surface: 
Distance 4.0 nm 3.0 nm 2.0 nm 1.0 nm 0.5 nm 0.25 nm 



Cation concentration 














(mol(-l-)L- 1 ) 


0.10 


0.12 


0.17 


0.35 


1.0 


2.0 


Anion concentration 














(mol(-)L-') 


0.10 


0.08 


0.06 


0.04 


0.01 


0.00 



QUESTIONS AND PROBLEMS 235 

Assuming that the excess of cations repotted for each increment represents the 
entire increment (e.g., that the cation concentration is 2.0 mol charge L _1 from 
the colloid surface to 0.375 nm from the surface, etc.), estimate the CEC for a 
colloid having 800 x 10 3 m 2 kg -1 of reactive surface (Ans. = 12.0 mmol charge 
kg" 1 ). 

2. Based on the data of Table 8.4, what proportion of the cross-sectional area of a 
cylindrical soil pore of radius 15 /zm is influenced by the electric double layer if 
monovalent ions predominate at a salt concentration of 10 _l M? 

3. If all water of an unsaturated soil at 20% water content is spread uniformly 
over 100 x 10 3 m 2 kg -1 of reactive surface, what proportion of that water is 
influenced by the electric double layer for the chemical conditions specified in 
Problem 2? 

4. A soil is equilibrated with a solution of SAR = 20. Based on the Gapon equa- 
tion, what would be its equilibrium exchangeable sodium percentage (ESP)? If 
the soil had instead been equilibrated with the same solution diluted fivefold with 
salt-free water, what would have been the corresponding SAR and ESP values? 

5. Generate a selectivity diagram similar to Figure 8.1 for two cations (A and B) 
having Kerr-type coefficients of 

(a) 0.5 

(b) 1.0 

(c) 2.0 

6. A vermiculitic surface soil has the ability to fix 25 mmol kg -1 of K + or NH4 . 
What rate of (NHU^SCXi or KC1 fertilizer (in kg ha -1 ) would be required to 
saturate this fixation capacity for a 30-cm depth of soil? 

7. The CEC is being estimated by the Okazaki, Smith, and Moodie procedure. If 5 g 
of soil retain 3 g 0. 1 M(+) index solution after centrifugation and decanting, and 
if the total index cation retained is subsequently determined to be 1.6 mmol(+), 
what is the CEC of the sample? Based on the anion distribution of Problem 1, 
what percentage error is contributed in this case by anion exclusion, if the soil 
has a reactive surface area of 200 x I0 3 m 2 kg - ' ? 

8. Based on layer lattice thickness estimates from Chapter 4, what is the relative 
attraction for a dehydrated K + ion residing directly on the mineral surface of a 
letrahedrally substituted 2:1 mineral when compared to an octahedrally substi- 
tuted 2:1 mineral? 

». For a mineral of CEC = 100 mmol(+) kg" 1 and surface area = 800 x 10 3 
m 2 kg -1 , saturated with a monovalent cation at a salt concentration of 0.001 M, 
use Fig. 8.8 to estimate the distance from the mineral surface to the midplane if 
Y b = 0.01, as well as the values of Y (= Zef/kT) at 0.5, 2, 5, 10, and 30 nm 
from the mineral surface. 



236 CATION RETENTION (EXCHANGE) IN SOILS 

10. Using the Y values from Problem 9, calculate the corresponding cation and anion 
concentration at the specified distances from the mineral surface. 

11. Verify the calculations of Table 8.5. 

12. Explain in your own words the differences between the Helmholtz, Guoy- 
Chapman, and Stern models of the double layer. 

13. Explain the valence dilution effect. 

14. For a CEC procedure that uses Na + as the index cation, H20/ethanol as the 
wash solvent, and Mg 2+ as the displacing cation, discuss the effect of each of 
the following on CEC measurements: 

(a) Hydrolysis due to excess washing 

(b) Presence of large amounts of lime or gypsum in the soil 

(c) Incomplete index-cation saturation 

(d) Precipitation of an insoluble Na + salt in the ethanol 

(e) Incomplete removal of the index cation by Mg + 

15. For a solution of SAR = 40 and a total salt concentration of 0.01 mol(+) L _l , 
calculate ESP iffe = 0.015. 



9 



ANION AND MOLECULAR 

RETENTION 



Soil particles remove anions and molecules from the soil solution, and release others 
to the soil solution, in varying degrees. The mechanisms of retention and release 
are electrostatic and chemical bonding. The mechanism of ion retention is actually 
exchange but the H + , OH~, and other ions released are usually unnoticed. These 
retention mechanisms are combinations of adsorption, absorption, precipitation, and 
solid solution mixing; distinguishing between the mechanisms is difficult. Soil-anion 
interaction varies from slight repulsion to weak to very strong attraction and retention 
(Table 9.1). 

Differences in retention between the groups is illustrated by Fig. 9.1. A soil natu- 
rally high in chloride and borate was leached successively with 40 pore volumes of 
water. (A pore volume of water is that amount which fills all the voids between the 
soil particles.) The C\~ concentration in the effluent was 1 100 mM initially and de- 
creased rapidly to almost zero after 5 pore volumes of water had passed through the 
soil. The H^BO^/f^BO^ concentration was about 15 mM initially and decreased 
much more slowly than C)~ . 

The leaching was stopped for 30 days and then was continued for 10 more pore 
volumes of water. The initial chloride concentration, the "rebound," was 0.5 mM 



Table 9.1. Anion and molecular interaction with soils 

Repelled to 

Weakly Retained Moderately Retained Strongly Retained 

N07,SO| - ,Se02" H 3 B0 3 ,H 2 BO^,F- H 2 P0 4 , HPO^-, H 2 S, HS~ 

HCOJ.CO^.CIO^ CrO^- H 2 AsO^, HAsCrj - , MoO^ - 

cr,Br-,r 

237 



238 



ANION AND MOLECULAR RETENTION 



TRAVER LOAM 
(i) Constant head 
(ii) Driving force = 9.8J/kg-m 
(iii) 0.033 M Ca(N0 3 ) 2 
(iv) 30 day Equilibration 




PORE VOLUME DISPLACEMENT 

FIGURE 9.1. Boron and Cl~ concentrations in successive pore volume displacements (PVD) 
of the Traver loam soil. The solid vertical line at PVD = 40 indicates an intervening 30-day, 
saturated storage period. F. J. Peryea, F. T. Dingham and J. D. Rhoades. 1985. Soil Sci. Soc. 
Am. J. 49:840. 



and the borate concentration was 0.3 mM. The shape of the chloride and borate 
curves were the same as the first leaching sequence: CI - decreased rapidly again and 
borate decreased more slowly. The relative amount of borate that was released to Ihe 
leaching solution during the 30-day incubation period was much greater than that of 
chloride. 

Chloride was not retained by the soils. The chloride increase in the second leach- 
ing sequence was due to CI - diffusion from pores that were stagnant during the first 
leaching sequence. The borate increase, on (he other hand, was due to a redistri- 
bution of available borate during the 30-day interval. Diffusion would account for 
only a tiny fraction of the second borate peak because the aqueous borate concentra- 
tion was so much less than the aqueous chloride concentration. The second borate 
peak was due to slow release of borate ions from retention sites in the soil, a redis- 
tribution of borate to reestablish the "equilibrium" between strongly adsorbed and 
weakly adsorbed borate that was disturbed when the first leaching sequence depleted 
the weakly held fraction. The strongly adsorbed sites may have been deeper in the 
soil's weathered surface layer. 



ANION AND MOLECULAR RETENTION 239 

Strongly retained ions such as phosphate are released to the soil solution in the 
same way as borate but at lower concentrations and at slower rates. Weakly retained 
ions, in contrast, reach an equilibrium or steady-state concentration between the soil 
surface and the soil solution very quickly. When added to soil suspensions in the 
laboratory, phosphate in the soil solution decreases rapidly at first and continues to 
decrease over periods of weeks in the laboratory and weeks to months in the field. 
The laboratory reaction goes faster because the mixing and contact between soil 
particles and the soil solution is more complete. 

Weak anion attraction and repulsion (anion exchange) in soils is primarily elec- 
trostatic and is similar to cation exchange. Anion exchange is rapid and reversible 
and the anion attraction is weak; chemical bonding is slower and stronger. Soils in 
Europe and North America are predominantly weakly to moderately weathered soils 
of neutral pH and have appreciable organic matter. In these soils the cation exchange 
capacity CEC greatly exceeds the anion exchange capacity (AEC). This preoccupa- 
tion with cation exchange goes back to the initial studies by Thompson and Way 
in England in 1850. Soil chemistry has reflected this geographical bias. Anion ex- 
change is important in Australia, New Zealand, and South Africa, where some soils 
are strongly weathered and have low organic matter contents and low pH; it occurs 
in European and North American soils, too. 

In strongly weathered soils of low pH and low soil organic matter, and in soils 
derived from volcanic parent material, the AEC can equal or exceed the CEC (Ta- 
ble 9.2). The predominance of AEC or CEC can change from one stratum to the next 
in the same soil as the pH and composition of the soil strata change. Many of these 
high- AEC soils are coincidentally in the Southern Hemisphere, but anion exchange 
is also significant in acidic and highly weathered soils of the southeastern United 
States, in European forest soils, and in Japan. 

One reason for the lesser interest in anion exchange may be that sulfate is the only 
macro-essential ion for plants that is retained to a significant extent as an exchange- 
able anion. Cation exchange, in contrast, covers four major cations: Ca, Mg, K, and 
Na. Each category in Table 9. 1 covers a range of retention. Among the "weakly 
retained" anions, sulfate and probably selenate are retained the strongest by soils, 
because of their divalent charge. Nitrate, chloride, and perchlorate are retained the 
weakest. Nitrate and chloride are indeed considered to be repelled, rather than re- 
tained, in predominantly negatively charged soils because their retention is so weak. 
As in cation exchange, anion retention depends on the size and charge of the hydrated 
ion and on the ability of the ion to covalently bond with the soil surface. 

The soil's retention of water-soluble cations and anions depends largely on col- 
loid and ion charge. The aluminosilicale layer-lattice minerals tend to dominate the 
clay fraction of temperate and arid region soils. These minerals are predominantly 
negatively charged, so their physical adsorption of cations exceeds their adsorption 
of anions. Many anions of interest, however, are weakly soluble because they form 
strong chemical bonds with the cations in soil clays. These bonds can overcome the 
electrostatic repulsion of the negative charge and lead to strong soil retention. 

Anion retention received little attention in North America until research on toxic 
wastes and on anionic pesticides demonstrated the importance of anion retention. 



240 



ANION AND MOLECULAR RETENTION 



Table 9.2. Charge characteristics of various soil orders 





Horizon 


PHsall 


Charge 


cmol c kg ' 


Soil Order 


Negative 


Positive 


Oxisol (Morais et al., 1976) 


A 


3.0 


3.9 


3.2 




A 


7.8 


11.6 


1.4 




B 


3.0 


2.5 


5.1 




B 


8.2 


5.3 


2.0 


Ullisol (Morais et al., 1976) 


A 


3.0 


1.0 


1.1 




A 


8.5 


5.0 


0.8 




B 


3.0 


2.5 


2.4 




B 


8.5 


3.8 


1.7 


Alfisol (Morais et al., 1976) 


A 


2.9 


3.8 


2.0 




A 


8.5 


14.0 


-1.6 




B 


2.9 


2.4 


5.5 




B 


8.5 


9.5 


1.9 


Andisol (Sumner et al., 1993) 


A 


4.6 


3.8 


2.6 




B 


5.1 


1.9 


4.2 


Oxisol (Sumner, 1963b) 


B 


3.5 


2.3 


5.1 




B 


8.2 


8.0 


1.0 



M E. Sumner. 1 998. Future Prospect* for Soil Chemistry, Soil Science Society of America, Madison Wl. 
Special Publication Number 55. 



Strongly weathered soils contain Al and Fe(III) hydroxyoxides whose negative 
charge is low and whose positive charge can be relatively high, especially at low 
pH. When acidic, these "variable charge" soils can retain more anions than cations. 
The amorphous weathering products of volcanic soils of Japan also exhibit variable 
charge. 

Molecular retention nowadays mostly refers to the widespread interest in the re- 
tention of organic pesticide molecules by soils. These molecules are mostly electri- 
cally neutral overall but have functional groups (PO^ - , Cl~, NO-7, etc.) that interact 
with organic and inorganic soil solids. Positively and negatively charged molecules 
behave somewhat like simple cations or anions. 

The soil's retention of uncharged molecules is largely independent of colloidal 
charge. For these substances, the soil is an inorganic matrix whose retention is based 
on the molecule's tendency to distribute between the gaseous, aqueous, and solid 
organic and inorganic soil phases. For organic molecules, this tendency depends on 
relative volatility, molecular weight, chemical composition, physical structure, sol- 
ubility in the soil solution, the soil's organic matter content, and to some extent the 
soil's surface area. If the molecule also contains functional groups of inorganic na- 
ture, such as R— CO, R— COOH, R— CHO, R— P0 4 , and R— NH 2 , soil retention 
increases. The R — CI unit of chlorinated hydrocarbons generally adds little to reten- 
tion except by increasing the molecule's mass and nonvolatility. 



ANION EXCHANGE 241 

The absorption of acid-forming gases (SO2, NO, NO2, HF, HC1) from the air 
increases with the molecule's water solubility and reactivity and increases with the 
soil's pH and base status. Weak bases such as NH3 are retained more strongly by 
acid soils. 

Soluble silica is ubiquitous in soil solutions, commonly at concentrations of about 
10 -4 M, or 2-5 mg Si L~\ and is present as SiOH4 rather than as an anion. It 
is less accurately described as silicic acid H4SiC>4 because it is a very weak acid, 
pK ~ 10"" 10 . Soluble silica already saturates the soil sorption sites so little is re- 
moved from solutions flowing through soil. Indeed, the course of soil weathering 
is the slow release of soluble silica to the soil solution. The chemistry of silica is 
dominated by very slow reaction rates and by the presence of many forms of silica 
and aluminosilicate minerals of varying aqueous solubility. The solubility of quartz 
(SiCh), chalcedony, chert, and other forms of Si02 is about 3 mg Si L~ ' . Amorphous 
and hydrated opal Si02 ■ H2O is soluble to the extent of 100 mg L _l . The range of 
equilibrium aluminosilicate solubility is broader, but equilibrium and silicate solubil- 
ity are rather incompatible terms. Particle size is at least as important a determinant 
of the soluble silica concentration in the soil solution. 

The anions of concern to agriculture include CI - , HCO^, NO^~, SO^ - , HPO^~, 
HiPO^, OH~, and F~. In addition, some micronutrients (H2BG^, MoO^ - , and 
HAsO^ - ) and heavy metals (CrO^ - ) exist as anions in soils, as do some pesticides, 
such as the dissociated phenoxyacetic acids (2,4,5-T and 2,4-D). Molecular species 
of interest include NH3, undissociated weak acids such as H3BO3 and F^SiCU, and 
the. undissociated forms of many pesticides (DDT, 2,4,5-T, ant 2,4-D.) The study of 
anionic and molecular retention by soils has been the subject of increasing research 
in recent years. 

The approach in this chapter is to describe various retention mechanisms and to 
cite examples of their involvement in the retention of specific anionic and molecular 
species. Several of the mechanisms are general and apply to many of the species 
listed above. Adsorption isotherms are also discussed because of their widespread 
use to describe anion and molecular retention by soils. 



9.1 ANION EXCHANGE 

Anions are attracted by positively charged sites on surfaces and repelled by negative 
charges. Layer silicates in the clay fraction of soils are mostly negatively charged 
so that anions tend to be slightly repelled electrostatically. Soils, however, contain 
solids, including the layer silicates, that also develop positive charges (often simulta- 
neously though in different locations). An anion approaching soil solids may thus be 
simultaneously repelled by negatively charged aluminosilicate surfaces and attracted 
to positive charges on clay edges, hydrous oxides, and allophane. 

If a dilute, neutral solution of KG is added to dry montmorillonite, the equilib- 
rium CI - concentration in the bulk soil solution will be greater than the Cl~ con- 
centration in the solution originally added to the clay. This phenomenon is observed 
whenever a salt solution is added to a dry colloid having no adsorbing capacity for 



242 ANION AND MOLECULAR RETENTION 

the anion at the prevailing pH. The process is called anion repulsion, or less ac- 
curately negative adsorption, and is due to anion repulsion from the diffuse double 
layer (DDL) surrounding charged colloid surfaces. An alternative explanation for the 
increased CI - concentration in the bulk solution is hydration of the montmorillonite 
with H2O, leaving less water for the salt. 

Factors affecting anion repulsion include (1) anion charge and concentration, 
(2) species of exchangeable cation, (3) pH, (4) presence of other anions, and (5) na- 
ture and charge of the colloid surface. Ions commonly exhibiting net anion repul- 
sion include Cl~, NO J", and SO^ _ . Anion repulsion, as moles repelled per unit 
area of solid surface, increases with anion charge (valence). If the negative charge 
of a soil colloid surface remains constant and if no other reactions take place, an- 
ions of higher charge are repelled more than anions of lower charge. Mattson found 
that anion repulsion in a Na-montmorillonite suspension increased in the order: 
Cl~ s» NO^ < SO4 - < Fe(CN)^". In soils that are dominated by Ca or other 
polyvalent cations, chemical reactions with the cations often change this purely elec- 
trostatic order. Increasing the anion concentration also increases the number of an- 
ions repelled, although the volume of the DDL from which the anions are excluded, 
the exclusion volume, decreases. 

Anything that affects the DDL also affects anion repulsion. Thus, the Cl~ exclu- 
sion volume of layer silicate suspensions increases in the order Ba 2+ < Ca 2+ < 
K + < Na + . The multiply charged and more tightly adsorbed cations better neutral- 
ize the negative charge and produce a more condensed double layer, so that a smaller 
number of anions is excluded and the exclusion volume is less. Lowering the pH 
decreases the soil's net negative charge and increases the positive charge, so anion 
repulsion decreases with soil pH. 

Anion repulsion also decreases when anions can be adsorbed by positively 
charged sites on soil colloids. Pretreatment of the colloids with highly charged 
and tightly adsorbed anions such as phosphate can mask the positive charges. These 
adsorbed anions present a negative surface to anions added later. Anion repulsion is 
then greater than in the absence of the tightly adsorbed anions. 

The greater the negative charge of the soil solids, the greater the anion repulsion. 
Montmorillonitic soils thus exhibit greater anion repulsion than do kaolinitic soils 
at all pH values, and especially at low pH, where kaolinite can develop a positive 
charge. Anion repulsion can have important consequences during solute transport 
through soils. When anions are excluded from some of the volume of water surround- 
ing soil particles, the anions can travel through the soil as a concentration bulge at 
the water front. The anions thus appear to travel faster through the soil than the water 
carrying them. 

Anions approaching positively charged sites on layer silicate or hydrous oxide 
minerals are attracted electrostatically in the same manner as cations are attracted 
to negatively charged soil colloids. The effects of ion concentration, valence, and 
complementary ion on the distribution of exchangeable anions are similar to the 
effects described for cations (Chapter 8). Electrostatically retained anions are said 
to be nonspecifically adsorbed. Figure 9.2a illustrates the nonspecific adsorption of 
CI - . The dotted line shows electrostatic attraction of a positively charged mineral 



STRONG ANION RETENTION 243 



■^ I +H- + a-= ^ \ to! 

^1 -"I 

<OH- |/2 /0H" I/2 

-^ j +N0 3 - = "\ I +W lb) 

<oB' OT ...a ^oS m «...no 3 - 

FIGURE 9.2. Nonspecific anion reactions at a solid/solution interface: (a) adsorption, (b) anion 
exchange. (After F. J. Hingston, R. J. Atkinson, A. M. Posner, and J. P. Quirk. 1967. Nature 
215:1459-1461.) 



surface site for the anion. The positive charge in this case is the result of surface 
protonation, which increases with soil acidity. Figure 9.2b shows the exchange of 
one nonspecifically adsorbed anion (NO J) for another (CI - ). Exchange equations 
similar to those developed for cation exchange describe such reactions because non- 
specifically adsorbed anions are in the solution adjacent to the solid surface and are 
readily exchangeable. 

The Cl~, NO^~, and SOlj - anions are considered to be nonspecifically adsorbed. 
Table 9.1 shows typical data for Cl~ and SO^ - adsorption by soils. The capacity 
of soils to adsorb anions increases with increasing acidity and is much greater for 
Hie kaolinilic soil, which has significant pH-dependent charge. At all pH values, the 
divalent SO^~ ion is adsorbed to a greater extent than the monovalent Cl~ ion, as 
would be expected on the basis of electrostatic attraction forces alone. 

For the montmorillonitic soil, where pH-dependent charge and thus positive 
charge are of minor importance, CI"" is adsorbed only slightly at low pH and not 
at all in the slightly acid to neutral pH range. Such data are typical of nonspecif- 
ically adsorbed anions. Even for kaolinite, and for soils containing considerable 
pH-dependent charge, anion adsorption is negligible at pH > 7. The generally 
negative charge of pH > 7 soils repels nonspecifically adsorbed anions. 

Chloride, nitrate, and sulfate are common and important anions in most soils and 
have been studied extensively. Chloride, in particular, is often used as an indicator 
of NOj mobility in soils, since CI - is not subject to the complicating biological 
reactions characteristic of NO^" . In most other respects, CI" behaves similarly to 
NO". 



9.2 STRONG ANION RETENTION 

Anions strongly retained by soils include P0 3 4 ~ , AsO^" , M0O4 "" , CrO;j"~ , and F~ . 
These anions are essential microelements for plants and animals and are present 
in trace concentrations in the solutions of native soils. Because the amounts and 
tenacity of soil retention of these ions is so much greater than CI, NO3, and others, 
this retention has been misnamed as specific adsorption. These anions are simply 



244 ANION AND MOLECULAR RETENTION 

water insoluble in the presence of the typical cations and colloids in soils. The state 
of these anions in the soil solution is a matter of great economic and environmental 
concern. Phosphate deficiency of agricultural crops is an ongoing global problem. A 
newer aspect is that these anions are being added to soils in fertilizers, agricultural 
wastes, fly ash from coal combustion, and municipal and industrial wastes. Initially, 
the additions increase the soil solution concentrations and plant availabilities of these 
anions. Since concentrations may reach levels that are appreciably greater than native 
levels, land disposal of such wastes has created public fear. 

Some concerns about the safety of waste disposal on soils may be exaggerated. 
Within days to several weeks, the plant availability and movement of many ions 
decrease sharply and are nearly indistinguishable from the native concentrations, if 
the wastes are well mixed with the soil. The native concentrations of all ions vary 
widely from soil to soil yet their concentrations in plants and groundwater are low. 
If the wastes are distributed widely and the soil is given some time to react with 
the wastes, the probability of contaminated food or water is very low. Concentrating 
wastes in "hazardous waste landfills" where we try to isolate wastes from the envi- 
ronment is well intentioned, but prevents soil mixing and may be, in the long run, a 
counterproductive method of dealing with wastes. 

The agricultural contribution to lake and stream contamination probably comes 
mostly from surface runoff of fertilized fields and from feed lots rather than from 
actual drainage water. Mixing, dilution, and lime can mitigate soil contamination 
problems. 

Anion removal from the soil solution is fast initially, but slows thereafter as the 
ions diffuse into the weathered and porous aqueous-solid phase on the surfaces of 
soil particles, increasing the diffusion pathlength. The diffusion also slows because 
the aqueous concentration, the driving force of the diffusion, is much less. Diffusion 
continues longer in the field than in the laboratory because the diffusion pathlengths 
from phosphate fertilizer granules to soil particles are much longer, and the water 
films on particle surfaces are thin. Phosphate retention in stirred suspensions in the 
laboratory reaches a steady state after several days. Phosphate from fertilizer gran- 
ules in the field can release phosphate to plants for weeks before the phosphate is 
slrongly adsorbed by the soil. 

A hydrous oxide system is amphoteric; that is, its surface charge varies from 
negative to neutral to positive, depending on the pH of the aqueous solution. An 
electrostatic approach explains the exchange properties of hydrous oxides for chlo- 
ride, sulfate, and other water-soluble anions. These surfaces can also interact chem- 
ically and strongly with weakly water-soluble anions. This gives Fe oxide- and Al 
oxide-dominated soils a much greater adsoiption capacity for these anions than that 
predicted from electroneutrality alone, that is, greater than the quantity of adsorbed 
anions required to neutralize the surface positive charge. Indeed, iron oxides and 
other oxides scavenge (remove) arsenate, phosphate, molybdate, and other anions 
from solution with high efficiency. 

Oxygen ions on a hydrous oxide surface can be replaced by oxyacid anions, such 
as phosphate, and by fluoride, which can enter into sixfold coordination with Al or 
Fe ions. This is known as ligand exchange, or anion penetration, for it takes place 



STRONG ANION RETENTION 245 



OH 2 + 0-5 



-1 + 1 



0H 2 



+0.5 



•Ct~ + NoF 



r-0.5 



0H 2 + 0- 5 



nO 



+ NoC«+ H2O 



(a) 



0H-°-5 
OH-05 



No +NaH 2 P(V ^ I 



-1.5' 



OH 



-0.5 



■2No + + H 2 (&) 



FIGURE 9.3. Specific anion reactions at a solid/solution surface: (a) Neutralization of positive 
charge, and (b) ionization of a proton of an adsorbed acid anion. (After F. J. Hingstort, R. J. Atkin- 
son, A. M. Posner, and J. P. Quirk. 1963. Trans. 9th Int. Cong. Soil Sci. 1:669-677.) 



within the crystal and renders the surfaces of oxides more negative. The negative 
charge arises when part of the liberated hydroxy! ions are neutralized by the forma- 
tion of water (Fig. 9.3). Ligand exchange can occur on surfaces initially carrying a 
net negative, positive, or neutral charge. This contrasts with nonspecific anion ad- 
sorption, which occurs only when the surface carries a net positive charge. Ligand 
exchange may explain why weak-acid anions show maximum adsorption at pH val- 
ues about equal to their pK values (Fig. 9.4). At pH = pK, both the amounts of 



300 1- 



^ 200 
o 

E 
£ 



IO0 



,'HF V 

■/ I 

/ » 
/ 1 
\ 
\ 
\ 




\ 
\ 
\ 
\ 




~4-_ ■•■<"'•. 




. ■ ■ v ^ 

\ 
\ 

H4S1O/J \ 

— 1 1 — a____) — l^_i — 1 — - 


\ \ 

-1 1 1 1 



6 7 8 
pH 



9 10 II 12 13 



FIGURE 9.4. The adsorption of three oxyacids and fluoride on geothite as a function of pH. HF, 
pK = 3.45; H 2 SeQ 3 , pK 2 = 8.35; H 3 P0 4 . pK^ = 2.12, pK 2 = 7.21, pK 3 = 12.67; H4S1O4, 
pKi = 9.66. (After F. J. Hingston, R. J. Atkinson, A. M. Posner, and J. P. Quirk. 1968. Trans. 9th 
Int. Cong. Soil Sci. 1:669-677.) 



246 ANION AND MOLECULAR RETENTION 

anion (dissociated acid) available for ligand exchange and the amounts of undissoci- 
ated acid capable of neutralizing liberated OH - are greatest. 

Fluoride ion is moderately adsorbed by soil minerals. Figure 9.4 shows fluoride 
adsorption by goethite (FeOOH). Fluoride adsoiption conforms to the ligand ex- 
change theory and is probably favored by the close similarity in size of F~ and OUT 
ions. In acid soils at equal anion concentrations, F~ adsorption predominates over 
that of other common anions. This makes F _ an effective desorbing agent for previ- 
ously adsorbed anions. 



9.2.1 Phosphate Reactions in Soils 

Phosphate is probably the most important example of specifically adsorbed anions. 
Many soils fix large quantities of phosphate by converting readily soluble phosphate 
to forms less available to plants. In terms of ligand exchange or anion penetration 
theory, phosphate adsorption on oxide surfaces can be explained by Fig. 9.5. Phos- 
phate replaces singly coordinated ("A-type") OH groups and then reorganizes into a 
very stable binuclear bridge between cations. 

In the laboratory, phosphate adsorption by layer silicates is rapid for a few hours 
and then continues more slowly for weeks. The initial rapid reaction can be envi- 
sioned as a combination of nonspecific adsorption and ligand exchange on mineral 
edges. The slower reaction probably consists of a complex combination of min- 
eral dissolution and precipitation of added phosphate with exchangeable cations or 
cations within the lattices. 

Low and Black showed that phosphate retention by kaolinite increased with time 
and phosphate concentration. Silica concentrations in the bulk solution increased 
simultaneously. The reaction was thought to be a two-stage reaction: 

Al 2 Si 2 5 (OH) 4 + 4H + + 2H 2 = 2Al(OH) 2+ + 2Si(OH) 4 (9.1) 

kaolinite 

AI(OH) 2+ + HPO 2 .- = A1P0 4 H 2 (9.2) 

variscile 

with the Al phosphate precipitating on the surface or phosphate tetrahedra substi- 
tuting for silicon tetrahedra. The more generally accepted explanation nowadays is 
the dissolution of Al 3+ as kaolinite breaks down, followed by the precipitation of 



,Fe-OH 
Ve-OH 



.Fe-OH 

♦h 2 po;= v 

N Fe-H 2 P0 4 



OH" 






FIGURE 9.5. Representation ot H 2 PO;j" penetration into an iron oxide surface and subsequent 
formation of a stable binuclear bridge. (After J. C. Ryden, J. K. Syers, and F. R. Harris. 1973. 
Adv. Agron. 25:1-15.) 



STRONG ANION RETENTION 



247 



Al phosphate. When hectorite (a 2:1 layer silicate in which Mg 2+ is the dominant 
octahedral cation rather than Al 3+ ) is substituted for kaolmite, much less phosphate 
is retained. This points out the importance of AI to phosphate retention. In acid soils, 
phosphate retention is often closely related to the amounts of extractable Fe 3+ and 
Al. 

Phosphate forms weakly soluble Fe 3+ and Al 3+ compounds at low pH, more sol- 
uble Ca 2+ and Mg 2+ compounds at pH values near neutrality, and difficultly soluble 
Ca 2+ compounds at higher pH. Lindsay and Moreno developed a solubility diagram 
for phosphate in a system containing variscite (AIPO4-H2O), strengite (FeP04-H20), 
fiuoroapatite (Caio<P04)6F2), hydroxyapatite (Caio(P04)fi(OH)2), octocaicium 
phosphate (Ca4H(P04>3), and dicalcium phosphate dihydrate (Q1HPO4 • 2HiO). 
The solubility diagram (Fig. 9.6) describes equilibrium phosphate precipitation 
reactions at various pH values, but slow kinetics have prevented the quantitative 
application of solubility data to soils. In addition, assumptions have to be made 
about the activities and the solid phases that control the activities of Ca 2+ , Fe 3+ , and 
Al 3+ in the soil solution. These assumptions limit the diagram to equilibrium, which 
can be approached but not reached in soils. Despite this shortcoming, Fig. 9.6 illus- 
trates the changes in phosphate minerals as phosphate fertilizers slowly transform to 
less-soluble states in soils. 

Figure 9.6 illustrates the relative stabilities of several phosphate compounds in 
soils of various pH values. Soil solution compositions can be plotted on the di- 
agram by measuring soil pH and soluble phosphate concentrations. Data above a 
compound's isotherm represent supersaturation with respect to the solid, indicating 
that the compound will precipitate. Data below the isotherm indicate urtdersaturation 
of phosphate in the soil solution with respect to that compound, so that the solid, if 



pH 2 P0 4 




FIGURE 9.6. Solubility diagram tor phosphorus compounds at 25° C and 5 x 10 -3 M calcium. 
(After W. L. Lindsay and E. C. Moreno. 1960. Soil Sci. Soc. Am. Proc. 24:177-182.) 



248 ANION AND MOLECULAR RETENTION 

present, would dissolve. Intersections of two isotherms represent solutions in equi- 
librium with both solids. 

Consider a soil of pH 4 to which soluble phosphate fertilizer is added, resulting 
in a phosphate potential, or PH2PO4 (= negative logarithm of the H2PO4 activity), 
of 0.5. This is point a, which falls on the dicalcium phosphate di hydrate isotherm 
and is at equilibrium with that solid phase. Point a is above the isotherms for fluo- 
roapatite, variscite, and strengite, indicating supersaturation of the aqueous solution 
with respect to these compounds. Fluoroapatite might tend to precipitate first, reduc- 
ing phosphate levels in solution (and increasing the value of PH2PO4). Assuming 
constant soil pH, PH2PO4 will tend to increase to the equilibrium line of fluoroap- 
atite (PH2PO4 = 2.0, point b). When PH2PO4 is greater than 0.5, the soil solution 
is undersaturated with respect to dicalcium phosphate dihydrate and it will dissolve, 
increasing the phosphate in solution once more. The phosphate concentration can be 
any value between a and b, depending on the rate of CaHP04 dissolution versus that 
of Ca5F(P04)3 precipitation. This precipitation-dissolution reaction will continue 
until all of the dicalcium phosphate dihydrate dissolves. The soil solution is then 
represented by point b, in equilibrium with fluoroapatite. The least soluble (most 
stable) compounds indicated on the diagram at acid pH are variscite and strengite. 
The soil solution at point b is highly supersaturated with respect to both of these 
compounds. As a result they should begin to precipitate immediately. The diagram 
predicts that both dicalcium phosphate dihydrate and fluoroapatite will eventually 
disappear to form variscite and/or strengite (points c and d). Either transformation 
results in a substantial reduction in soluble phosphate concentrations compared to 
those of the initially fertilized soil (point a). Large quantities of phosphate should 
thus be fixed as insoluble Fe 3 " 1 " and Al phosphates in acid soils. 

The diagram also indicates that phosphate should precipitate in basic soils as one 
of several Ca phosphates, the least soluble of which are hydroxy- and fluoroapatite. 
Variscite and strengite are too soluble to exist under basic conditions, and they should 
not form in basic soils. Both variscite and strengite, in fact, would be good phosphate 
fertilizers in alkaline soils because of their solubility in basic soils, if they were 
applied as finely ground materials. Calcium phosphate ore ("rock phosphate," mostly 
hydroxy- and fluoroapatite) is similarly effective in acid soils. Rock phosphate is 
treated with sulfuric acid to make "superphosphate," nominally CaHP04; treatment 
with phosphoric acid yields "triple superphosphate," nominally Ca(H2PC>4)2. Both 
superphosphate and triple superphosphate are more soluble than rock phosphate and 
make phosphate more immediately available when added to soils at any pH. 

Figure 9.6 explains the observations that (1) phosphate is fixed in large amounts 
as iron and aluminum phosphates in acid soils, where Fe 3 ' 1 " and Al 3+ activities are 
high; (2) calcium fixes phosphate similarly in basic soils, where Ca 2+ activity is 
high; and (3) maximum amounts of phosphate are available at slightly acid to neutral 
pH where the solubilities of Fe, Al, and Ca phosphates are highest simultaneously. 

For most soils, the various mechanisms responsible for phosphate retention are 
nearly impossible to separate. Phosphate is retained by a multistage process, proba- 
bly involving several of the mechanisms described above as well as other unknown 
reactions. Even carefully designed experiments are often confounded by reactions 



STRONG ANION RETENTION 



249 



other than those intended to be studied. Precipitation is especially difficult to elim- 
inate as a mechanism; all of the specific adsoiption mechanisms can be viewed as 
modified precipitation reactions. 

The mechanisms of phosphate retention by soil organic matter are not known, 
but are believed to be important in maintaining plant-adequate levels of phosphate 
in the soil solution. Inositol hexaphosphale and possibly other organic phosphate 
compounds apparently are retained in soils by precipitation reactions. Many com- 
mon, water-soluble, organic phosphate compounds become nonextractabie to water 
at almost the same rate as, and to the same extent as, dissolved inorganic phos- 
phates. Thus, although organic phosphate is reported to leach from soils, a large 
proportion of it appears lo move attached to particulate matter rather than as dis- 
solved phosphate. Retention mechanisms for organic phosphate include (1) sorption 
through orthophosphate groups to Fe and Al oxides by mechanisms similar to those 
for inorganic phosphate, and (2) sorption by interaction of the organic portion of the 
phosphate ester with organic or inorganic soil components. 

At current rates of fertilizer usage, we have about a 200-year supply of good-grade 
phosphate ore worldwide. The importance of phosphate fertilization to agriculture 
and the relatively short supply of phosphate ore have led to many, as yet unsuc- 
cessful, attempts to increase the low availability of the large amounts of phosphate 
present in almost all soils. These attempts include leaching the soil with silicate, 
which might replace phosphate that is strongly retained by Fe, Al, and aluminosili- 
cates; creating polymeric phosphate fertilizers (pyro- and metaphosphates) instead of 
the normal ortho (monomelic) forms; and breeding plant varieties that can better uti- 
lize soil phosphate. To overcome the high cost of shipping phosphate per unit P and 
of spreading it on rugged terrain, New Zealanders went so far as to propose spreading 
elemental P from airplanes. White P, however, is a dangerous explosive and the safe, 
polymeric red and black P forms are too insoluble and unreactive in soils. Unlocking 
native phosphate would be an important step in achieving sustainable agriculture. 

The "breakthrough cun'es" of Fig. 9.7 summarize the net effects of repulsion and 
specific adsorption on the relative adsorption of anions by soils. Solutions contain- 
ing the anions at initial concentration Co were added to soil columns. The effluent 
concentration is C. The volume of water is expanded as pore volumes added to the 



1.0 



o 



0.5 





cr / 






-4. 


/so/" 


/h 2 bo 3 " 
■ i j — i — 1_ 


i i — ■ — ■ — i — 



4 6 8 10 
PORE VOLUMES 



12 14 16 



FIGURE 9.7. Representative breakthrough curves of anions weakly, moderately, and strongly 
retained by soils. 



250 



ANION AND MOLECULAR RETENTION 



soil column. The G~ and NOJ solutions flowed through the soil columns almost 
as quickly as the water. The other anions were delayed because of soil adsorption. 
Sulfate and phosphate retention increased with iron and aluminum hydroxyoxide 
and allophone contents. The long-term capacity of most soils to adsorb phosphate is 
orders of magnitude greater than the amounts of phosphate added as fertilizer. 



9.3 MOLECULAR RETENTION 

A solute in water need not be initially charged to be retained by soils. Molecules in 
the soil solution can become charged and then be adsorbed as cations or anions. They 
may also remain nonionic and adsorb as a consequence of polarity that produces 
localized charge within the molecule. 

Molecules such as NH3, amino acids, and protein can protonate (add H + ) in acid 
solutions and be adsorbed as cations on negatively charged soil solids: 



B + H+ = BH + 



(9.3) 



where B is a weakly basic molecule. The tendency of a molecule to protonate is 
characterized by its pK a : 



K a = 



(H+)(B) 
(BH+) 



pK- d = pH + log 



(BH+) 
(B) 



(9.4) 



The greater the pAT a of a basic molecule, the greater is its tendency to protonate. 
Important molecules that protonate include the i-triazine and .s-triazole herbicides 
and ammonia. Their p£ a values are given in Table 9.3. 

When soil pH > pK, weak-acid anions are adsorbed by positively charged sites 
on Fe and Al oxides or layer silicate edges. The weak acids (high pK a values) include 



Table 9.3. p/C a values of some molecular species 



Species 




P^a"- >' 


Reaction 


,s-Trazine 


(atrazine) 


1.68 


BH+ = B + H + 


s-Trazole (amitrole) 


4.17 


BH+ = B + H + 


2,4-D 




2.80 


R— COOll = R— COO - + H+ 


2,4,5-T 




3.46 


R— COOH = R— COO - + H + 


H 2 C0 3 




6.37 


H 2 C0 3 = HCO^ + H f 


H3BO3 




9.14 


H3BO3 = H 2 BO^ 4- H+ 


NH 3 




9.26 


NH| + OH" = NH 3 + HOH 


H 4 Si0 4 




9.66 


H 4 Si0 4 = H3S1O- + H+ 



"Organic pAf„ reprotluced from Pesticides in Soil and Water, 1974, p. 47. 

^Inorganic p/f a reprinted vvich permission from Handbook Chemistry and Physics. 50th ed., 1969-1970. 
Copyright, The Chemical Rubber Co., CRC Press, Inc. 



MOLECULAR RETENTION 251 

H3BO3 and H4SiC>4 (or Si(OH) 4 ). These acids remain uncharged in most agricultural 
soils and do not form anionic or cationic bonds. Important weak acids that dissociate 
to form anionic bonds with soil include the phenoxyacetic acids (2,4-D and 2,4,5-T) 
and carbonic acid (H2CCH). Their p£ a values are also given in Table 9.2. 

Molecules that do not protonate or deprotonate to become charged species can still 
be adsorbed on soil by hydrogen bonding and van der Waals attraction. The hydrogen 
bond is a dipole-dipole interaction in which H + bridges between two electronegative 
atoms. The hydrogen is held by a weak electrostatic bond to one electronegative 
atom and by a stronger covalent bond to the other. The functional groups of the 
soil's solid phase that are capable of hydrogen bonding include the 2 ~ on silicate 
surfaces, edge hydroxyls, and the carboxyl, hydroxyl, and amino groups of organic 
matter. Individual hydrogen bonds are relatively weak, but many polar molecules 
(particularly pesticides) have numerous sites capable of hydrogen bonding with soils, 
especially with soil organic matter. The summation of many hydrogen bonds results 
in strong retention of, for example, carbaryl and carbamate insecticides. 

Many organic molecules, although uncharged and without apparent hydrogen 
bonding, are nonetheless strongly retained by soils. The intense interest in this 
phenomenon stems from possibility of movement of pesticides and other organic 
molecules in soils to groundwater. A less obvious phenomenon is the soil's adsorp- 
tion of organic molecules from the atmosphere. Uncharged molecules have been 
adsorbed from the atmosphere and produced in the soil by organic decay since the 
earth was formed, yet the groundwater and atmosphere are remarkably free of them. 
This is partly due to the strong retention of organic molecules by soils, a second 
reason is the active degradation of organic substances by soil microorganisms. 

Soil retention of uncharged molecules is often described as van der Waals attrac- 
tion, which is a way of saying that the retention mechanism is unknown or poorly 
understood. In the 1 9th century, van der Waals modified the ideal gas law to account 
for the attraction between gas molecules, without knowing the nature of attraction. 
Charge-induced dipole interactions and dipole-induced dipole interactions are the 
forces thought to be involved. The van der Waals attractions are weak and short- 
ranged. They are additive, and each atom of the molecule and its adsorbent contribute 
to the total bond energy. Such forces operate in all adsorbent-adsorbate relationships 
but appear to be the principal forces of adsorption for nonpolar molecules such as 
DDT and N2. The electrostatic forces of charged species overshadow van der Waals 
attractive forces. 

Molecular retention involves no charges and therefore requires no strict l-l ex- 
change between the soil and the soil solution. Ion retention requires exchange to 
maintain charge neutrality. The amount of molecular retention, however, is limited 
by the number of exposed sorption sites, or by the amount of sorbing surface and 
material, in the soil. 

Percolating solutions containing organic molecules pass through an intricate net- 
work of soil pores. Organic molecules tend to be nonpolar and to prefer an environ- 
ment less polar than that of the highly polar water. If some other less polar phase is 
present, such as soil surfaces and especially SOM surfaces, the organic molecules are 
in effect forced out of the aqueous phase onto organic-coated soil surfaces. The SOM 



252 ANION AND MOLECULAR RETENTION 

also attracts organic molecules by providing a phase into which they can "dissolve" 
or form a solid solution. That action helps to purify contaminated water or gas flow- 
ing through the soil. The soil's adsorption capacity is small for organic molecules 
but is continually renewed by microbial decay of the adsorbed molecules. 

The separation of organic molecules out of the soil solution onto the solid phase 
is called partitioning. The ratio of a molecule's concentrations in the water and SOM 
phases is a constant, the partition coefficient K\y. 

concentrationi (s() ii) 

ad = — — : (?■•>) 

concentrauonn()«,ii solution) 

and is a measure of the relative solubility of the molecule in both phases. For sub- 
stances that are only slightly soluble in water, the values of Kn are very large. Diben- 
zothiophene ((C6H5>2C4H4S) is weakly water soluble and its soil/water K& value is 
1 1 000. Passing dibenzothiophene-contaminated water through soil containing or- 
ganic matter greatly depletes the water of this compound. 

The extent to which an organic compound partitions out of water onto soil is deter- 
mined by physical-chemical properties of both the soil and the compound. The soil's 
organic matter content is the single best characteristic for estimating the amount of 
soil adsorption of pesticides and other organic molecules. The partition, or sorption, 
coefficient of the organic molecule Koc (equal to K D /SOM) is rather independent 
of soil type. This suggests that SOM is the principal soil component responsible for 
pesticide sorption and that the role of SOM is similar in different soils. 

The Koc value is correlated to physical-chemical properties of the organic 
molecules. One such easily determined property is the partition coefficient A'ow of 
a molecule between octanol and water replicates fairly well the partitioning between 
soil and the soil solution. The correlation of Xow to Koc m soils is 

log Koc = -0-99 log A- 0W - 0.34 (9.6) 

The Koc is a first approximation of a pesticide's mobility in soil from readily avail- 
able pesticide and soil properties. 

Partitioning by this hydrophobic adsorption explains why soils retain organic 
molecules, but direct soil-organic interaction occurs also. Dry soils retain organic 
molecules more strongly and in greater amounts than do wet soils. Differences be- 
tween organic molecules with respect to soil retention become more obvious when 
the competition with water for soil surfaces is absent. For a gas passing through a 
dry soil, the soil/gas partition coefficient of methane is about the same as that for 
dinitrogen (N2) and helium. Methane flows as easily through soil as the unreactive 
N9 and He molecules. The partition coefficient increases exponentially with molecu- 
lar weight to about 10 5 for gaseous octane (Fig. 9.8). Unsaturated double bonds and 
aromatic ring structures increase retention slightly, and alcohol, aldehyde, and acid 
functional groups in the gas molecule increase its soil retention greatly. Presumably, 
nitrogen, phosphate, and sulfur functional groups also increase retention. Those dis- 



MOLECULAR RETENTION 



253 



UNSATURATED _ 
HYDflOCAA80NS~ 




SYMBOLS 

A METHANE 
> ETHANE 
C PROPANE 
S ailTANE 
E HEXANE 
F HEPTANE 

OCTANE 
N ETHYLENE 

1 ACETYLENE 
J CYCIOHEXANE 
K tSOeuTTLENE 
L ISOOCTANE 
M PENTANC 
N (CNZCNE 



4 6 

CARBON NUMBER 



FIGURE 9.8. Retention of various hydrocarbons by a dry soil at 1 5° C. Vg, the retention volume, 
is closely related to the soil/gas partition coefficient Kq. (From H. L. Bohn et al. 1980. J. Environ. 
Qual. 4:563.) 



tinctions, however, are less obvious in the presence of water because it competes with 
the gases for adsorption sites. For more polar and water-soluble organic molecules, 
direct adsorption by inorganic soil surfaces and dissolution into the soil solution are 
also important. 

The adsorption of a particular uncharged species can rarely be identified with only 
one mechanism, although the dominant mechanism can often be inferred. Thus, care 
should be taken in extrapolating data from, for example, a weakly basic herbicide 
to other weakly basic herbicides. The individual properties of the molecule, such 
as (1) chemical character, shape, and configuration, (2) acidity or basicity, (3) wa- 
ter solubility, (4) charge distribution, (5) polarity, (6) size, and (7) polarizability, all 
influence molecular adsorption by soil. 

As with cations and anions, soil interaction with molecules happens only if the 
substance contacts soil particles. Surface spreading or burial is insufficient; the 
wastes must be mixed with the soil to react with it. The sensational Love Canal 
case, for example, involved the burial of organic liquids in 55-gallon (215-L) drums 
stacked in shallow trenches underground. Had the organic liquids been mixed and 
allowed to interact with the soil, the leakage and movement after the thin steel 
drums corroded might never have happened, and it certainly would have been less 
severe. 



254 



ANION AND MOLECULAR RETENTION 



9.4 ADSORPTION ISOTHERMS 

Adsorption isotherms describe solute adsorption by solids at constant temperature 
and pressure. An adsorption isotherm shows the amount of adsorbate sorbed as a 
function of its equilibrium concentration. A variety of isotherm shapes are possible, 
depending on the affinity of the adsorbent for the adsorbate (Fig. 9.9). 

To generate adsorption data, a known amount of adsorbate in aqueous solution 
is mixed with a known amount of adsorbent. At equilibrium, the amount of adsor- 
bate removed from solution is assumed to be adsorbed. Secondary reactions (such as 
precipitation) must be eliminated or corrected for. Precipitation is indicated in some 
cases by a rapid increase in apparent adsorption (disappearance from solution) with 
a small change in solution concentration. Three equations are commonly used to de- 
scribe adsorption: the Langmuir, Freundlich, and Brunauer-Emmett-Teller (BET) 
equations. 



Monoloyer 









Solution Concentration 
(c) 



FIGURE 9.9. Typical adsorption isotherms described by the (a) Langmuir, (b) Freundlich, and 
(c) BET equations. 



ADSORPTION ISOTHERMS 255 

The Langmuir equation was initially derived for the adsorption of gases by solids. 
The derivation was based on three assumptions: (1) a constant energy of adsorption 
that is independent of the extent of surface coverage (i.e., a homogeneous surface); 

(2) adsorption on specific sites, with no interaction between adsorbate molecules; and 

(3) maximum adsorption equal to a complete monomolecular layer on all reactive 
adsorbent surfaces (Fig. 9.8a). A common form of the Langmuir equation is 

*- = -«™- (9.7) 

m \+KC 

where x/m is the weight of adsorbate per unit weight of adsorbent, K is a constant 
related to the binding strength, b is the maximum amount of adsorbate that can be ad- 
sorbed (i.e., a complete monomolecular layer), and C is the adsorbate concentration. 
Rearranging Eq. 9.7 yields the more convenient linear form 

x/m Kb b 

If adsorption conforms to the Langmuir model, plotting C/(x/m) versus C yields a 
straight line with a slope l/b and intercept 1/Kb. The Langmuir constant K is the 
quotient of the slope \/b and intercept 1/ Kb. 

Equation 9.7 assumes constant free energy of adsorption on the surface, a situa- 
tion that rarely occurs in nature. Instead, the energy of adsorption tends to decrease 
with increasing surface coverage. The interaction with already -adsorbed molecules 
increases with increasing surface coverage. The net effect is that the two phenomena 
tend to compensate for each other, yielding a relatively constant energy of adsorp- 
tion. In systems where the energy of adsorption is not constant, the Langmuir equa- 
tion may still describe adsorption over a portion of the adsorption range, since the 
variation in energy of adsorption over such a range can be small if only one type of 
bonding site or mechanism predominates. 

A true adsorption maximum, however, is rarely observed. Precipitation reactions 
can exhibit Langmuir-type behavior. If only a limited quantity of a solute that pre- 
cipitates is present, a Langmuir isotherm can result as the solute increases, that is, 
a "sorption maximum" occurs. This behavior is found at low solute concentrations, 
where no precipitation occurs until the solute's solubility product is reached. 

An advantage of using the Langmuir equation for describing adsorption is that it 
defines an adsorption limit on a given array of sites that meet the Langmuir model's 
criteria. This limit has been used to estimate the adsorption capacity of soils for 
phosphate and various herbicides. Comparing such capacities can also suggest ad- 
sorption mechanisms. Unfortunately, the adsorption maxima usually do not occur, 
instead adsorption continues but at ever-decreasing amounts. 

If data fail to conform to the Langmuir equation, the Freundlich equation often 
fits the data successfully: 

- = KC l ' n (9.9) 

m 



256 



ANION AND MOLECULAR RETENTION 



where K and n are empirical constants and the other terms are defined above. The 
equation was originally empirical, without a theoretical foundation. It implies that 
the energy of adsorption decreases logarithmically as the fraction of covered surface 
increases, similar to the solid solution ideas. The Freundlich equation can be derived 
theoretically by assuming that the decrease in energy of adsorption with increas- 
ing surface coverage is due to surface heterogeneity. The degree of heterogeneity is 
unknown in most adsorption studies, and both the Langmuir and Freundlich equa- 
tions are better thought of as empirical curve fitting of adsorption data, rather than 
describing the actual mechanism of adsorption. 
The linear form of the Freundlich equation is 



x 1 
log — = - log C + log K 
in n 



(9.10) 



The frequent good fit of adsorption data to the Freundlich equation is influenced 
by the insensitivity of log-log plots and by the flexibility afforded curve fitting by 
the two empirical constants K and n. This flexibility does not guarantee accuracy, 
however, if the data are extrapolated beyond the experimental range. The Freundlich 
equation has the further limitation that it does not predict a maximum adsorption ca- 
pacity, however mythical the adsorption maximum may be. Despite its shortcomings, 
the Freundlich equation is a common adsorption equation and is included in several 
models for predicting pesticide behavior in soil. 



APPENDIX 9.1 MULTISITE AND MULTILAYER ADSORPTION 

A number of recent studies involving the adsorption of solutes from solution by min- 
eral surfaces have resulted in data suggesting multiple-site adsorption. That is, sev- 
eral different arrays of sites are postulated, each of which fulfills the requirements of 
the Langmuir model. For example, Fig. 9. 10 shows data for phosphate adsorption on 



c 

x/m 




FIGURE 9.10. Phosphorus adsorption data plotted according to the Langmuir equation. (After 
J. K. Syers, M. G. Browman, G. W. Smillie, and R. B. Corey. 1973. Soil Sci. Soc. Am. Proc. 
37:358-363.) 



MULTISITE AND MULTILAYER ADSORPTION 257 

soil plotted according to the traditional linear Langmuir form. The data suggest two 
sets of sites, each with its own binding strength and adsorption maximum. Alterna- 
tively, the data suggest two mechanisms of adsorption on similar sites. The adsorp- 
tion curve is resolved by dividing the curve into several straight-line components. 
For Fig. 9.10, only two straight lines are needed. The form of the Langmuir equation 
for two-site adsorption is 

^ =hl _ x J^i +h2 - x Jan ( 9.,,) 

m K X C l K 2 C ' 

where subscripts 1 and 2 refer to regions (or mechanisms) 1 and 2. The adsorption 
maximum for the soil is the sum b\ plus b^- This form of the Langmuir equation can 
be used to describe adsorption sites for species whose total adsorption appears to be 
the sum of both a high-energy and a low-energy component. Such a distribution of 
adsorption energies is reasonably well established for soil phosphate. To fit adsorp- 
tion data more closely to the Langmuir model, even three- and four-site models have 
been invoked. 

In addition to multisite adsorption, many gases and vapors adsorbed by solids do 
not produce a typical monolayer-type adsoiption isotherm (Fig. 9.9a), but rather pro- 
duce an isotherm indicating multilayer adsoiption (Fig. 9.9c). An equation that treats 
multilayer adsorption is the BET equation, named after developers Brunauer, Em- 
mett, and Teller. Multilayer adsoiption is characteristic of physical or van der Waals 
attraction. It often proceeds with no apparent limit, since multilayer adsorption 
merges directly into capillary condensation as the vapor pressure of the adsorbate 
approaches its saturation value. 

The BET equation has been used to determine the surface area of solids from gas 
adsoiption data. The equation not only predicts the shape of the adsorption isotherm, 
but also gives the volume of gas V m required to form a monolayer. The BET equation 
has the form 

A = ^N a A m (9.12) 

where A is the surface area of the adsorbent, V m is the volume of gas adsorbed, 
Vb is the molar volume of adsorbate gas (22.4 liters at 25° C), N d is Avogadro's 
number, and A m is the cross-sectional area of the adsorbate molecule. Surface area 
determinations are often based on Nj adsorption by the solid at —195° C. For N2, 
,4 m = 16.2 x I0~ 20 m 2 . The linear form of the BET equation is 

' + ^^ <M3) 



V(Po-P) V m C V m CP 

where P is the equilibrium pressure at which a volume V of gas is adsorbed, Po is the 
saturation pressure of the gas, and C is a constant related to the heat of adsorption of 
the gas on the solid. If a plot of Pj V(Po - P) is a straight line, the effective surface 
area of the solid can be calculated after C has been determined, either from the slope 
of the line (C - l)V m C, or from the intercept, \/V m C. 



258 ANION AND MOLECULAR RETENTION 

The BET equation has been applied to ion adsorption from soil solutions, although 
the extended Langmuir equation (Eq. 9.11) would seem to apply as well. The BET 
equation has also been used to study the adsorption of pesticides having relatively 
high vapor pressures. 

An alternative explanation of retention by adsorption is to consider the substance 
being held by solid-state mixing on the surfaces of soil particles (see Chapter 3). 
The advantage to the solid-state mixing concept is that the quantitative interpreta- 
tion is based on the same thermodynamics as ions in aqueous solutions. It provides a 
plausible explanation of why the retention energy progressively decreases as the ad- 
sorbate concentration increases, why soils retain ions at different energies, and why 
soil particles retain substances much more strongly than their own pure minerals 
while releasing other ions as the soil weathers. 

Retention by solid solution cannot be differentiated from adsorption on the basis 
of experiment. They are simply different explanations for the same phenomenon. 
The idea of adsorption sites can be reconciled by considering that those areas are 
where solid-state mixing, because of the solid's surface structure and composition, 
is most likely. 



BIBLIOGRAPHY 

Edzwald, J. K., D. C. Toensing, and M. C. Leung. 1976. Phosphate adsorption reactions with 

clay minerals. Environ. Set. Tech. 20:485-490. 
Guenzi, W. E. (ed.). 1974. Pesticides in Soil and Water. Soil Science Society of America, 

Madison, WI. 
Mott, C. J. B. 1981. Anion and ligand exchange. In The Chemistry of Soil Processes (D. J. 

Greenland and M. H. B. Hayes, eds.). Wiley, New York, Ch. 5. 



QUESTIONS AND PROBLEMS 

1. Distinguish between specific and nonspecific reactions of anions soils. Give ex- 
amples of anions that tend to be specifically and nonspecifically reactive in soils. 

2. What are the forces acting on an anion as it approaches a layer silicate? How will 
these forces vary from soil to soil? What is the dominant for acting on anions in 
most agricultural soils? 

3. How do the following factors affect anion repulsion: 

(a) Anion charge 

(b) Anion concentration 

(c) Exchangeable cation 

(d) SoilpH 

(e) Other anions 



QUESTIONS AND PROBLEMS 259 

4. Are all anions adsorbed alike in soils? If not, explain the differences, giving 
examples of each reaction type. 

5. What reactions are responsible for the fixation of phosphate in acid and basic 
soils? 

6. How are molecular species such as N2 and NH3 retained by soils? 

7. Certain mechanisms are active in the retention of all species (cationic, anionic, 
and molecular), while other mechanisms are active in the retention of only cer- 
tain species. Explain, giving examples of species retained predominantly by each 
specific mechanism. 

8. Refer to Fig. 9.5. Describe the sequence of events if a soil of pH 7.5 were fertil- 
ized to a pH 2 P0 4 level of 2. 

9. Given the data below, determine if the adsorption of 2,4,5-T conforms to the 
Langmuir or the Freundlich models and determine the appropriate adsorption 
parameters (K, n, b). You may need to restrict your attention to a limited con- 
centration range. 



Initial Solution 


Final Solution 








Concentration 


Concentration 


Volui 


me of Solution 


Weight of Soil 


(mgL- 1 ) 


(mgL" 1 ) 




(mL) 


(g) 


5 


3 




10 


5 


to 


6 




10 


5 


25 


15 




10 


5 


50 


30 




10 


5 


100 


70 




10 


5 



10. Explain in your own words the peaks and inflection points of Fig. 9.3. 

11. Maximum phosphate availability in soils tends to occur around pH 6 to 6.5. 
Explain why in terms of Fig. 9.5. 

12. Based on Table 9.3, predict the relative mobility of (a) s-triazole, (b) H3BO3, 
and (c) 2,4-D in pH 4.5, 7.0, and 8.5 soils. 



10 



ACID SOILS 



Rainfall over a large portion of the earth's surface exceeds evapotranspiration for 
much of the year, and soil leaching results. The leaching gradually removes soluble 
salts, more readily soluble soil minerals, and bases (nonacidic cations such as Ca 2+ , 
Mg 2+ , K + , and Na + ). Consequently, the leached surface soil becomes slightly to 
moderately acid, although the subsoil may remain neutral or alkaline. As weathering 
proceeds, even acidic components are leached from the soil. At this stage, the surface 
soil pH, and ultimately the pH of the entire profile, once more approaches neutrality. 
Only iron and aluminium oxides, and some of the trace metal oxides that are also 
highly resistant to weathering, remain in the soil from the original parent material. 

Local highly acidic conditions can also arise when mine wastes containing iron 
pyrite (FeS?) and other sulfides are exposed to the air. The sulfide oxidizes to H2SO4 
and Fe(OH)3. Acidities of pH 2 or lower are not uncommon in these soil solutions. 
Extremely acid soils also result from drainage of marine floodplains containing high- 
sulfide sediments ("cat clays" or Acid Sulfate soils), which oxidize to H2SO4 upon 
exposure. 

Crop fertilization can also produce substantial soil acidity. Continued use of am- 
monia fertilizers can lead to acidic soil conditions by the microbially mediated reac- 
tion 

NH+ + 20 2 = NOJ + 2H+ + H 2 (10.1) 

Less acidity is generated from NH4NO3 per unit of nitrogen than from (NH4)2S04, 
because only half of the nitrogen in NH4NO3 can be further oxidized. The H3PO4 
released by dissolving phosphate fertilizer granules can lead to pH values near the 
granule as low as pH 1.5. The H3PO4 is rapidly neutralized by soils, but the acidic 
reaction products may remain to influence soil properties. Despite considerable pub- 
licity about acid rain, the rate of soil acidification due to acid rain (containing H2SO4 

260 



ACID SOILS 



261 



and HNO3 as a result of human activities) is normally severalfold lower than that 
from the use of ammoniacal and phosphate fertilizers. In nonagricultural soils, acid 
rain is a relatively greater factor in soil acidification, partly because these soils are 
not limed to overcome acidity, as are agricultural soils. 

Finally, acidity may be produced by plant residues or organic wastes decompos- 
ing under somewhat reducing conditions into organic acids. This is of particular im- 
portance in many forest soils. The organic acids account in part for the dissolution 
and movement of Fe, Al, and Mn through the soil beneath many forest litter layers. 
Chelation or complexation by soluble organic molecules also contributes to cation 
transport through the soil under such conditions. 

The chemical behavior and properties of acid soils and the diagnosis and amelio- 
ration of their adverse effects are the main subjects of this chapter. The problem of 
plant growth in acid soils is treated only lightly. Excellent books by Black and by 
Pearson and Adams treat this aspect of the subject in more detail. 

The soil chemistry literature of the 1930s, 1940s, and early 1950s contains re- 
ports of numerous studies on the properties of hydrogen-saturated soils and clays. 
One aim was to predict the amount of lime needed to counteract soil acidity. A com- 
mon approach was to titrate the soil potentiometrically (as a function of pH) with a 
base. Figure 10.1 shows typical curves for the potentioinetric titration of acids, and 
of titrating an acid montmorillonite suspension, with a strong base such as NaOH. 
Curve 1 is typical of strong acids such as HO, and is also typical of a freshly prepared 
acidic clay suspension. The pH remains relatively constant until nearly all of the acid 
is neutralized. Then the pH rises rapidly until it is determined by the concentration 
of the added base. 

Curves 2 and 3 are NaOH titration curves of polyprotonated weak acids, such 
as H3PO4, of titrating an acid clay suspension prepared slowly by dialysis, of clay 
suspensions prepared rapidly by newer methods and then aged for a few days, and 



12 


- 








u 










10 


- 








9 










8 


- 








7 

PH 6 

5 




3 


*Z 


y 1 


4 


_ • 








3 








1 Fresh suspension 








2 Dried from water 


2 
1 


L 


1 


1 


3 Dried 1rom alcohol 
I I I I I 



20 40 60 80 100 120 
me NaOH/lOOg CLAY 



140 160 



FIGURE 10.1. Potentiometric titration of montmorillonite suspensions after treatment with 
NaCI-HCI solution and H-resin. (From D. G. Aldrich and J. R. Buchanan. 1958. Soil Sci. Soc. 
Am. Proc. 22:281-286.) 



262 ACID SOILS 

of field soils. Here the pH increases more or less continuously as base is added, with 
an occasional plateau corresponding to weak-acid groups having pK values near 
the plateau pH. The intermediate plateaus in this figure, for example, correspond to 
weak-acid pK values of 5.5 to 6. 

In 1947, Chernov of the Soviet Union summarized the results of many studies 
about the nature and properties of acid soils and clays. He recognized that hydrogen- 
saturated minerals were highly unstable, and that they rapidly broke down to release 
Al, Mg, and Fe from within their lattices. He suggested that most hydrogen-saturated 
soil materials were in reality saturated primarily with Al 3+ and Fe 3+ . His work was 
supported in the United States during the early 1950s by Jenny, Coleman, and others. 
They showed that weak-acid properties commonly attributed to "hydrogen" clays 
(e.g., curves 2 and 3 in Fig. 10.1) were in reality the result of partial or complete 
saturation of exchange sites with weakly acidic Al ions. For these two curves, 40 to 
50% of the cation exchange capacity (CEC) was actually occupied by exchangeable 
Al. Hydrogen clays that were analyzed immediately after rapid preparation behaved 
much more like strong acids. For curve 1 in Fig. 10.1, for instance, less than 10% of 
the clay's CEC was Al saturated. Results such as these suggested that acid soil clays 
were behaving like Al and Fe(III) clays rather than H clays. 



1 0.1 INSTABILITY OF HYDROGEN SOILS AND CLAYS 

Hydrogen soils and clays prepared by strong-acid leaching or dialysis decompose 
rapidly to Al- and Fe(III)-saturated materials. The half-life for the temperature- 
dependent decomposition is only a few hours for many minerals (Table 10.1). 
Hydrogen-saturated Utah bentonite (smectite) was half converted to the correspond- 
ing Al-saturated form after 18 hours at 30° C, and three-fourths converted after 36 
hours. Corresponding times at 60° C were 2. 1 and 4.2 hours. Wyoming (Volclay) 
bentonite was stable nearly three times longer than the Utah bentonite when H satu- 
rated, and kaolinite was stable nearly seven times longer. The reaction can be slowed 
markedly by storage at temperatures near freezing, but truly H-saturated soils or 
clays must still be studied as rapidly as possible. 



Table 10.1. Rates of decomposition of hydrogen-saturated layer silicates, showing the 
time for one-half of the exchangeable hydrogen to the loss at various temperatures 9 

Rates of Decomposition (Half-time) (min) 
Layer Silicate 30° C 50° C 60° C 70° C 80° C 

Montmorillonite 1080 260 125 60 32 

(Utah Bentonite) 
Montmorillonite — — 340 — 86 

(Volclay bentonite) 
Kaolinite — — 850 — 194 

" N. T. Coleman and D. Craig. 1961. Soil Sci. 91:14-18. 



HYDROLYZED ALUMINIUM IONS 



263 



Vermiculite is particularly susceptible to decomposition under acid conditions. 
This susceptibility is one explanation for the relatively low vermiculite contents of 
slightly acid surface soils from arid and semiarid regions, when compared with corre- 
sponding subsoils. Another explanation for low vermiculite contents of surface soils 
is K cycling by plants, with the subsequent conversion of vermiculite particles to 
their collapsed (micaceous) equivalents. 



1 0.2 HYDROLYZED ALUMINIUM IONS 

The chemical behavior of acid soils and minerals is intimately linked to the aque- 
ous solution chemistry of aluminium. Aluminium hydvolyz.es to monomelic and 
polymeric hydroxyaluminium complexes made up of Al(OH) 2+ and Al(OH)J. Ulti- 
mately Al precipitates as solid-phase gibbsite (Al(OHb) when the solubility product 
of this mineral is exceeded. The hydrolysis reaction*; of the monomers are 

Al(H 2 0)^+ + H 2 = Al(OH)(H 2 0)f + H 3 + 
Al(OH)(H 2 0)^ + + H 2 = Al(OH) 2 (H 2 0)| + H 3 + (10.2) 

AKOH) 2 (H 2 4 ) + +H 2 = Ai(OH) 3 (H 2 0)^ + H 3 + 
Al(OH) 3 (H 2 3 ) + H 2 = Al(OH) 4 (H 2 0)2 + HjO + 

Each reaction is driven to the right by the consumption of H + (hydronium, H-jO" 1 ") 
ions by reacting with hydroxyl ions. Successive hydrolysis reactions are associated 
with increasing pH. The distribution of Al ions with pH is shown in Fig. 10.2. For 
convenience, the H 2 ligands have been omitted from the formulas. Such diagrams 




FIGURE 10.2. Relative distribution and average charge of the soluble aluminum species as a 
function of pH, ionicstrength = 0.1 M. (From G. Marion et al. 1976. SoilSci. 121:76-82.) 



264 ACID SOILS 

identify the Al hydrolysis species al various pH values and show their relative con- 
tribution to total soluble Al. The Al(OH) 2+ ion is of minor importance and exists 
over only a narrow pH range. The Al 3+ ion is predominant below pH 4.7, Al(OH)2 
between pH 4.7 and 6.5, Al(OH)^ between pH 6.5 and 8, and AKOH)^ above pH 
8. The Al(OH)j" ion occurs only at pH values above those common to soils. Solid- 
phase Al(OH)3 precipitates throughout the pH range covered, whenever its solubility 
product is exceeded. The total concentration of soluble aluminium is strongly pH de- 
pendent and is minimal at about pH 7. 

The hydrolysis reactions of Eq. 10.2 liberate H + and lower the solution pH unless 
OH~ is present. This stepwise production of hydrogen ions is similar to that which 
occurs during the dissociation of polyprotonated acids. It is the primary reason why 
early workers attributed weak-acid properties to acid soils and clays. 

Monomeric hexaquoaluminium (A1(H 2 0);? + ) is exchangeable, although its triva- 
lent charge results in a strong retention or preference by many soil colloids. The 
hydrolyzed Al ions rapidly polymerize to form large, multicharged units. The poly- 
merization is enhanced by soil colloid surfaces. Hydroxyl groups are shared by ad- 
jacent Al ions to produce polymers of the general formula (Al (OH) v (H 2 0)£ ~ )„ , 
where « is the average number of Al ions per polymer. These multicharged poly- 
mers polymerize (age) further with time. The polymers are strongly retained by soil 
colloid surfaces and behave as if they are virtually nonexchangeable. This is prob- 
ably because sufficient numbers of exchanging ions are rarely present at any time 
or place to exchange them. An exception to the general rule of nonexchangeability 
for hydroxy aluminium polymers occurs when previously formed polymers are ad- 
sorbed on expanded monlmorillonite interlayers. In this rare circumstance, the poly- 
mer may attach to the monlmorillonite surface in only a few places, and hence remain 
exchangeable when exposed to sufficient numbers of replacing cations. 

Retention of positively charged and virtually nonexchangeable hydroxy alu- 
minium polymers lowers the net negative charge of soil colloids. Thus, formation 
of hydroxy aluminium polymers on the surface of inorganic soil colloids decreases 
the cation exchange capacity (CEC) of the colloids. Raising the pH decreases the 
positive charge on the polymers (Eq. 10.2 and Fig. 10.2) and increases the CEC 
of the mixture. This is an important source of pH-dependenl charge for inorganic 
soil colloids. Conversely, lowering the pH of soils containing large quantities of 
adsorbed hydroxy aluminium polymers decreases the soil CEC, by increasing the 
positive charge on the polymers. In some cases it may result in zero to positive 
charge. 

Iron hydrolysis is similar to that of aluminium. The pK of the first step of Fe(III) 
hydrolysis, 

Fe(H 2 0) 3+ + H 2 = Fe(OH)(H 2 0)* + + H 3 (10.3) 

is close to 3, while that for Al(H 2 0)g + is 5. The Fe(III) ion is a stronger acid and its 
acidity is buffered by the Al hydrolysis reactions. Most of the soil's large reserve of 
Al would have to react before pH could decrease to the point where Fe hydrolysis 
could control soil pH. So Al 3+ is the primary ion of concern in acid soils. 



CLASSIFICATION AND DETERMINATION OF SOIL ACIDITY 265 

Hydroxy aluminium and hydroxy iron polymers also can adsorb anions with con- 
current release of hydroxyl ions. The pH increase due to this anion exchange can 
be masked, however, by the simultaneous hydrolysis of desorbed aluminium ions 
(Eq. 10.2). Adsorption of multicharged anions can also decrease the net positive 
charge on hydroxy aluminium or hydroxy iron polymers, and thus increase the net 
negative charge of the soil-polymer mixture. The anion adsorption capacity of soils 
decreases with increasing pH and becomes virtually zero for all anions except phos- 
phate and arsenate at pH values greater than 5.5 or 6. 

Hydroxy Al and hydroxy Fe polymers can be held between the lattices of expand- 
ing soil minerals, preventing collapse of these lattices as water is removed during 
drying or freezing. Swelling of dried soil materials may also be restricted, since wa- 
ter is less able to enter between mineral sheets once the minerals have collapsed. As 
little as 1/16 coverage by a hydroxy aluminium or hydroxy iron monolayer can sta- 
bilize soil minerals against shrinking and swelling. The resultant minerals are termed 
intergrades: chlorite-vermiculite and chlorite-montmorillonite intergrades are par- 
ticularly common. Although intergradc minerals commonly are described as having 
chlorite-like structures, the interlayer generally consists of either hydroxy aluminium 
or hydroxy iron compounds, rather than the hydroxy magnesium of most geologic 
chlorites. The polymeric materials in acid soils and days also tend to exist as in- 
termittent islands between the interlayers of expanding minerals, rather than as the 
continuous sheets typical of traditional chlorites. 

Once present as interlayer material, hydroxy aluminium or hydroxy iron polymers 
can be removed only by fairly complete neutralization of their charge to form un- 
charged Al(OH>3 and FeOOH or by acidification to produce monomeric (and hence 
exchangeable) Al 3+ or Fe 3+ . Upon neutralization to form the aluminium or iron 
hydroxides, which is complete at about pH 8 for aluminium, the hydroxides and the 
negatively charged silicate layer lattices are no longer electrostatically attracted. Fur- 
thermore, particles of intergrade minerals are generally smaller than the free gibbsite 
(Al(OH)s) and goethite (FeOOH) particles of soils. The higher surface energy of the 
smaller particles causes them to dissolve from mineral surfaces and to reprecipitate 
as part of the larger particles. At still higher pH values, the dominant soluble species 
are negatively charged Al(OH)J, Fe(OH)^ , and Fe(OH)^". Hence, they are repelled 
by negatively charged layer silicates. 



10.3 CLASSIFICATION AND DETERMINATION OF SOIL ACIDITY 

Various approaches have been used to classify the components of soil acidity. As a 
carryover from the titration curves used to characterize soil acidity in earlier stud- 
ies, a common category is titratable acidity or total acidity. This is the quantity of a 
strong base (NaOH or Ca(OH)2) required to raise soil pH to a predetermined level. 
Time, method of stirring, and period between additions of base must be specified 
because the neutralization of soil acidity is highly dependent on reaction conditions. 
The values are also meaningless unless the initial and final pH values are specified, 
because more base is consumed if the reaction is carried out over a wider pH range. 



266 ACID SOILS 

For example, titration curve 3 in Fig. 10.1 required approximately 200 mmol hy- 
droxy! kg -1 of clay to raise the pH from 4 to 5, and another 200 mmol hydroxy 1 
kg -1 of clay to raise the pH from 5 to 6. 

Common endpoints of such titrations are pH 7 or pH 8.2, although soils in the field 
are rarely limed above pH 6 or 6.5. The value 8.2 was chosen historically because it 
approximates the pH of soil containing free CaCC>3 in equilibrium with the normal 
CO2 content (0.0003 mol fraction) of the atmosphere. This pH also corresponds 
closely with the pH of complete neutralization of soil hydroxy aluminium com- 
pounds. The pH 8.2 is conveniently maintained by Mehlich's BaCh-triethanolamine 
extraction technique. 

The titration process, if carried out so slowly that the reaction is fairly complete 
following each addition of base, does not distinguish between exchangeable and vir- 
tually nonexchangeable components. Hence, titratable acidity is only a measure of 
the total acidity neutralized during the experimental technique employed. The titrat- 
able or total acidity is nonetheless useful for determining the lime requirement of 
acid soils. 

Further classification of acid soils includes the distinction between exchange- 
able and nonexchangeable acidity. Exchangeable acidity is that exchanged by an 
unbuffered neutral salt solution, such as 1 M KC1 or NaCl. Nonexchangeable acidity 
also has been more ponderously termed "titratable but nonexchangeable acidity." It 
includes hydroxyl-consuming reactions such as neutralization of hydroxy aluminium 
polymers on soil surfaces: 

X— Al(OH) 2+ + OH" = X— Al(OH)+ (10.4) 

neutralization of protons from weakly acidic organic functional groups: 

R— COOH + OH" = R— COO" + H 2 (10.5) 

(creating additional pH-dependent charge or CEC); and displacement of adsorbed 
anions: 

X— A1(0H)(H 2 P0 4 ) + OH" = X— Al(OH) 2 + H 2 POJ (10.6) 

Exchangeable acidity consists of monomeric aluminium and exchangeable hydro- 
gen. The exchangeable aluminium and exchangeable hydrogen can be roughly di- 
vided by soil pH, for exchangeable hydrogen normally is present in measurable 
quantities only at pH values less than 4. Hence, exchangeable hydrogen is of concern 
only for extremely acidic materials, such as mine spoils or acid sulfate soils from ma- 
rine floodplains, organic acids from decomposing soil organic matter, and from acid 
rain. Exchangeable aluminium normally occurs in significant amounts only at soil 
pH values less than about 5.5. In the range of pH 5.5 to 7, hydroxy aluminium poly- 
mers predominate among acidic soil components, exchangeable acidity is virtually 
absent, and only nonexchangeable and titratable acidity are present in measurable 
quantities. Significant quantities of such acidity from weakly acidic R — COOH and 



CLASSIFICATION AND DETERMINATION OF SOIL ACIDITY 



267 



R — OH groups of soil organic matter, and from incompletely neutralized hydroxy 
aluminium polymers can be present in soils at pH > 7. 

Although exchangeable acidity is essentially absent above pH 5.5, some direct 
proton exchange by weaker carboxylic groups and most of the phenolic groups on 
SOM, as well as some of the weakly acidic protons on soil mineral edges, may still 
occur above this pH. H + production from such groups is relatively minor in most 
agricultural soils. Some ambiguity also exists when the CEC of organic colloids is 
neutralized by difficultly exchangeable Al 3+ and Fe 3+ . These ions may react during 
relatively rapid titrations as tilratable but nonexchangeable acidity. 

More precise separation of acidity into its components requires a separate alu- 
minium (and possibly iron) determination on the neutral salt extract, or a conduc- 
timetric titration (Fig. 10.3). The electrical conductance changes rapidly with the 
degree of neutralization when relatively mobile ions, such as hydrogen or hydroxy I, 
predominate in solution. It changes little when immobile ions, such as aluminium 
or hydroxy aluminium, dominate. In Fig. 10.3, the conductance (reciprocal of resis- 
tance) decreases as the H + concentration decreases during the change from pH 3 to 
5. The resistant then remains unchanged, near pH 5, as the added OH~ precipitates 
aluminium. When all the Al 3+ ions have been titrated to Al(OH)3, the conductance 
increases again as more OH - is added. The similarity of such a titration curve to the 
titration curves of acid soils is one type of evidence that led to the conclusion that 
acid soils are aluminium soils rather than true weak acids. 

The complexity of soil acidity emphasizes the need to specify the final pH of a 
titration if the components of soil acidity are to be classified meaningfully. Classi- 
fication also should include specifying the initial pH, which can indicate the most 
probable forms of acidity present. 



500 



400 - 



300 - 




ZOO - 



O.A Q& 

mmol NaOH 



FIGURE 10.3. Potentiometric (•) and resistance (o) curves for titrating mixtures of HCI plus 
AICI 3 with NaOH. (From P. F. Low. 1955. Soil Sci. Soc. Am. Proc. 19:135-139.) 



268 ACID SOILS 

10.4 SOIL PH MEASUREMENTS 

Soil pH measurements can be ambiguous. Two factors that affect soil pH measure- 
ments are the soil-solution ratio and the salt concentration. Increasing either factor 
normally decreases the measured soil pH because H and Al cations on or near soil 
colloid surfaces can be displaced by exchange with soluble cations. Once displaced 
into solution, the Al ions can hydrolyze (Eq. 1 0.2) and further lower the pH. Prefer- 
ential retention of hydroxy aluminium polymers by soil colloids drives the hydrolysis 
reactions further toward completion and leads to lower pH. Increasing the neutral salt 
concentration to 0. 1 or 1 M can lower the measured soil pH as much as 0.5 to 1 .5 
units, compared to soil pH measured in distilled water suspensions. 

Because of the cation distribution in the diffuse double layer (DDL), and possibly 
because of higher concentrations of hydrogen near weakly ionized organic groups 
and mineral edges, the hydrogen ion concentration near soil colloid surfaces ap- 
pears to be 100 to 1000 times greater than in the bulk solution. Such concentrations 
have been substantiated for dry clays by measuring the infrared spectra of adsorbed 
organic acids, and for wet clays by the pH-dependent reaction rates of adsorbed en- 
zymes. The greater acidity near soil colloid surfaces should be kept in mind whenever 
adsorption mechanisms are being postulated for weakly acidic organic molecules, in- 
cluding many of the common pesticides. 

In soil suspensions at low salt concentrations, extraneous (junction) potentials 
can affect pH readings (Appendix 10.1). The most plausible explanation for junction 
potentials is that K ions in the KC1 bridge of the reference electrode diffuse more 
rapidly, and CI ions less rapidly, when negatively charged soil particles are near the 
bridge. One answer is to place the reference electrode in the clear supernatant solu- 
tion and the glass electrode in the settled clay suspension (where H + is concentrated) 
to obtain valid soil pH measurements. Junction potentials are essentially eliminated 
at salt concentrations greater than 0.01 M. 

10.5 PERCENT BASE SATURATION 

A soil parameter of historical importance is the percent base saturation, defined as 

v- 100 

percent base saturation = > (exchangeable Ca, Mg, Na, K) x (10.7) 

at pH 7 or 8.2. The pH used for this measurement and for CEC determinations must 
be specified whenever this concept is used. As an example, consider a soil of pH 
5 that has 5 mmol(+) of exchangeable bases (Ca, Mg, K, and Na), 1 mmol(H-) of 
exchangeable acidity, and a CEC of 80 at pH 7 and of 100 at pH 8.2. The percent 
base saturation of this soil is 62% based on the CEC at pH 7, 50% based on the CEC 
at pH 8.2, and 83% al the native soil pH. If organically chelated or polymeric forms 
of aluminium are present, the basic cations that are displaced may vary somewhat 
with the pH of the extracting solution. This is probably because of competition for 
exchange sites from small, pH-dependent quantities of displaced aluminium. 



LIME REQUIREMENT 269 

In the early literature dealing with soil acidity, soils were characterized by their 
percent base saturation at specified pH levels. Soils with low percent base saturation 
values were considered to be dominated by kaolinite and hydrous oxide minerals, 
but soils of high percent base saturation were considered to be dominated by 2:1- 
type minerals, such as mommorillonfte, vermiculite, chlorite, and the micas. Base 
saturation is a criterion of soil taxonomy in the U.S. soil classification scheme. Fifty 
percent base saturation (based on soil CEC at pH 7) is one criterion for distinguish- 
ing between mollic epipedons (dark, high organic horizons) and their umbric (low 
organic) counterparts. 

Unfortunately, percent base saturation is as much a measure of the pH-dependent 
charge of soils as it is of the actual percentage of cation exchange sites occupied 
by exchangeable bases. The denominator includes any additional charge (CEC) gen- 
erated SOM and hydrous oxide-mineral complexes between the initial soil pH and 
the reference. pH (7 or 8.2). Since neither exchangeable aluminium nor exchangeable 
hydrogen is appreciable above pH 5.5, the CEC above this pH should be 100% base 
saturated. However, soils in the pH range 5.5 to 7 or 8.2 generally still have measured 
base saturations well below 100%. Such base saturation values are particularly low 
for minerals that have a high proportion of pH-dependent charge, such as kaolinite. 

Although imprecise, the percent base saturation is still useful for soil genesis 
and classification puiposes and for empirical liming recommendations. For exam- 
ple, each unit rise in soil pH was related to a 20 to 30% increase in base saturation 
for field soils in Virginia. From the standpoint of soil chemical properties and re- 
actions, however, the base saturation is more correctly an acidity index or liming 
index. In addition, the degree of nonbase saturation is more meaningful if separated 
into exchangeable acidity and pH-dependent charge. 



10.6 LIME REQUIREMENT 

A major problem of managing acid soils is to estimate the quantity of lime required 
to raise the soil pH to a certain level. As shown in Table 10.2, plant species vary 
considerably in their response to soil pH. Such data must be interpreted carefully. In 
this case, the nonlegumes benefited from nitrogen fixed by legumes in the rotation. 
Much of the pH response may actually be the pH response of nitrogen fixation by 
the legumt-Rhizobium pair. 

The most theoretically satisfying way to estimate the lime requirement of acid 
soils is to measure the quantity of base required to raise soil pH to a specified level. To 
be realistic the titration must be slow enough for the added base to react completely 
with the soil. Both exchangeable and titratable acidity will be neutralized during the 
titration. 

Ten mmoI(+) kg" ' OH - consumed during the titration is equivalent to 4.5 tonnes 
pure CaCC>3 (ha-30 cm) - " 1 of field soil. The mol(+) equivalent weight of CaC03 is 
100 in this case because only one OH - is produced per CaCC>3 molecule, in the 
normal pH range of limed acid soils. Although many people still regard the primary 
effect of lime to be the provision of adequate soil calcium, its main value is really to 



270 ACID SOILS 



Table 10.2. Yield of crops grown in corn, small grain, legume, 
or timothy rotation at different soil pH values 8 





Average Relative Yield at pH Indicated 


Crop 


4.7 


5.0 


5.7 


6.8 


7.5 


Sweet clover 





2 


49 


89 


100 


Alfalfa 


2 


9 


42 


100 


100 


Red Clover 


12 


21 


53 


98 


100 


Alsike clover 


13 


27 


72 


100 


95 


Mammoth clover 


16 


29 


69 


100 


99 


Timothy 


31 


47 


66 


100 


95 


Barley 





23 


80 


95 


100 


Corn 


34 


73 


80 


100 


93 


Wheat 


68 


76 


89 


100 


99 


Oats 


77 


93 


99 


98 


100 



"From Ohio Agric. Expt. Sta. 1938. Ohio Auric. Expl. Sta. Special Circular 
53. 



provide hydroxyl ions: 

CaC0 3 + H 2 = Ca 2+ + HCO^ + OH - (10.8) 

Increased quantities of soluble and exchangeable Ca and Mg are by-products of lim- 
ing, though their greater amounts may be beneficial to plants, such as legumes, hav- 
ing high Ca requirements. 

A hydroxyl ion is also consumed during the displacement of adsorbed anions as 
the soil pH is raised. This effect is not normally a major one, but contributes to field 
lime requirements of several metric tons per hectare for some highly acid Piedmont 
soils from the southeastern United States. 

Field liming reactions are generally incomplete, because of incomplete mixing, 
and require considerable time. The reaction rale varies inversely with pH, limestone 
particle size, and solubility of the liming agent. Hence, the laboratory lime require- 
ment value is often further multiplied by a conversion factor to better estimate the 
amount of lime needed to achieve a given field pH. 

The titration of individual soil samples is impractical for soil-testing purposes, 
because of the time and experimental precision required. Such titration is also highly 
dependent on the time allowed for each increment of base to react with the soil 
(Fig. 10.4). The usual procedure is to add a pH buffer solution to the soil, measure 
the amount of buffer consumed or the resulting pH of the soil-buffer suspension, 
and calibrate results with field lime requirements for similar soils from the same 
geographical area. 

If the soil is leached with buffer solution until the soil pH equals that of the buffer, 
titrating the remaining buffer capacity of the solution measures the soil acidity that 
must be neutralized to produce a soil pH equal to that of the buffer. A more rapid 
method is add buffer solution to soil without attempting to bring the final pH of the 



LIME REQUIREMENT 



271 




pH recorded after : 
G 30 minutes 
o 4 hours 
• 24 hours 
A 48 hours 



0.2 0.3 04 

mmol NaOH 



0.5 



0.6 



FIGURE 10.4. NaOH titration curves for vermiculitic Greenfield soil as influenced by time of the 
titration interval. (From A. L. Page et al. 1 965. Soil Sci. Soc. Am. Proc. 29:246-250.) 



mixture to the initial pH of the buffer. The pH of the soil-buffer suspension indicates 
the degree of soil acidity present and, after field calibration, indicates the quantity 
of lime required to raise the soil pH to a certain level. For example, a pH change 
of 0.1 unit from the initial buffer pH might correspond to 1.0 Mg limestone ha -1 . 
If a soil has an initial pH of 5.5, the buffer solution has an initial pH of 6.8, and 
the final mixture has a pH of 6.3, then the lime requirement for this soil would be 
1.0 x 5, or 5.0 metric tons ha -1 . The pH values of the final mixtures are calibrated 
against those of similar samples that have been leached with buffer solutions, or that 
have been titrated to specified pH levels, to obtain more precise estimates of the lime 
requirement. 

Estimates of soil texture and measurements of initial soil pH for similar soils 
from a rather homogeneous geographical area can provide a simpler but less precise 
estimate of soil lime requirements. Such techniques must be calibrated against one of 
the more precise lime requirement methods to accurately estimate the amount of lime 
required. Different limed-soil pH values each require a separate calibration curve. 

To achieve maximum crop production, soil pH must be raised to the optimum 
level for the crop in question. Little is gained by raising the pH to still higher levels. 
The growth increase from each successive increment of lime diminishes, but the cost 
of adding the increment remains the same. Although complete equilibration with 
lime may not occur until pH 8 or 8.2, acid soils are rarely limed above pH 6 or 6.5. 

Plant growth in strongly weathered soils can be hampered by acidic subsoils. Sur- 
face application and mixing by plowing and discing are ineffective in treating the 
subsoils. Lime diffuses no more than 10 cm even 10 years after application. Adding 



272 ACID SOILS 

gypsum and lime at the same amounts (ca. 5-10 Mg ha" 1 each) on the surface im- 
proves plant growth and considerably improves root growth and penetration of the 
treatment into the subsoil. Gypsum (CaS04 • 2H2O) is more soluble than CaCCb. 
The sulfate anion is thought to penetrate to the subsoil, saturate the positively charged 
clays, raise pH by 0.4 units, and reduce Al toxicity. 



10.7 ALUMINIUM AND MANGANESE TOXICITY 

Many plants grow poorly in acid soils. Early workers supposed that this was a conse- 
quence either of hydrogen ion toxicity or of Ca and Mg deficiencies. The soil acidity 
must be greater than about pH 3, however, before the H + concentration itself is toxic 
to most plant species. Although the components of acidity are emphasized in acid 
soils, the major exchangeable cations are Ca, Mg, and to a lesser extent K in soils of 
pH > 4.5 to 5. 

Plant growth problems associated with poor root penetration into acid subsoils are 
frequently associated with high plant availability of Al or M, which are toxic to most 
plants. Aluminium restricts or stops root growth at solution concentrations as low as 
1 mg L _l . Plants tolerate higher levels of soluble manganese, but reducing condi- 
tions in flooded or periodically inundated acid soils can result in soluble manganese 
concentrations as high as 100 mg L _1 . 



10.8 PH AND MACRONUTRIENTS 

The effects of low pH on plant growth are generally caused by increases of toxic 
ions, or decreases of essentia) ions, in the soil solution. Such effects can also arise 
from nutritional imbalances because the concentrations can increase or decrease as 
soil acidity changes. 

Although effects of pH on plant nutrient levels in soils are complicated and inter- 
related, some generalizations are possible. Plants able to utilize ammonium forms of 
nitrogen have a considerable advantage in acid soils, because nitrification (microbial 
oxidation of ammonium to nitrate) is slow below pH 5.5. Ammonium ions may ac- 
cumulate in acid forest soils, because the microbes that mineralize organic nitrogen 
to ammonia are less dependent on soil pH than are the nitrifying organisms. 

Another important facet of nitrogen availability in acid soils is the pH dependence 
of ammonium-ion fixation between the lattices of expanding layer-silicate minerals. 
Such fixation generally decreases with increasing soil pH. Although the mechanism 
for this pH effect is incompletely understood, the decrease may be due to "islands" of 
hydroxy aluminium and hydroxy iron polymers, which prevent the complete collapse 
of mineral lattices and hence decrease NHj fixation. 

The availability of soil phosphate is highly pH dependent and, as with nitrogen, 
is only partially understood. The main mechanism for phosphate fixation (decreased 
availability) under acid conditions appears to be the precipitation of highly insoluble 
iron and aluminium phosphates. Phosphate availability also tends to decrease at high 



PH AND MICRONUTRIENTS 273 

soil pH, because of precipitation on insoluble calcium phosphate compounds. The 
pH range of greatest phosphate availability is about 6 to 7 for most agricultural soils. 

Liming acid soils can increase or decrease potassium availability. Decreased K 
availability can be attributed to increased K fixation in limed soils, similar to ammo- 
nium fixation. Liming can increase K availability where the soil in its native state 
may have insufficient nutrient cations for plant growth. This would be typical of 
sandy soils or of highly weathered tropical soils. Insufficient exchange sites may be 
present in such soils to retain K and other nutrient cations against leaching. The in- 
creased soil CEC upon liming retains greater quantities of fertilizer K within the root 
zone and also retains it longer. 

The change in ion-exchange specificity with pH (Chapter 8) can also cause op- 
posing trends of K availability in limed soils. Increased availability can be attributed 
to greater quantities of K in the soil solution, because of Ca replacement of K in 
the DDL of the soil's colloids. Decreased K availability after liming can be due to 
greater quantities of K leaching from limed soils. 



10.9 PH AND MICRONUTRIENTS 

The micronutrients of major interest to soil chemistry because of plant deficiencies 
are boron, manganese, iron, cobalt, copper, zinc, and molybdenum. Other ions — 
chromium, nickel, cadmium, mercury, and lead — behave similarly in soils but the 
problems are usually plant toxicity. The availability of most of the micronutrient 
and toxic ions increases with increasing soil acidity. Those present as anions — 
molybdenum, chromium, and boron — differ in that their availability generally de- 
creases with increasing acidity. 

Acid soils generally provide sufficient micronutrients, occasionally even toxic 
amounts, to plants or to animals grazing on those plants. Because of the small quan- 
tities of micronutrients required for plant growth, adequate amounts can be taken up 
from small portions of the root zone, if such regions are sufficiently acidic. In basic 
soils the acidity from fertilizers or from small quantities of elemental sulfur or sulfu- 
ric acid added to a portion of the root zone may provide adequate micronutrients to 
plants. 

Molybdenum is unique among the micronutrients because it is less available to 
plants at low pH. Occasionally, the harmful effect of soil acidity on leguminous 
plants seems to be caused by Mo deficiency rather than by Al toxicity (Table 10.3). 
In this case of a soil in the foothills of the Cascade Mountains in Oregon, the normal 
fertility program of P, K, B, and S gave only low to moderate alfalfa yields. Liming 
at a rate of 5000 kg ha ' gave high yields in all cases, as did only 0.5 kg ha -1 of Mo 
on the unlimed soil. Molybdenum is required for nitrogen fixation by legumes. 

In addition to pH effects on the availability of individual ions, various nutrients 
often interact with respect to their effects on plant growth. Some such interactions 
may arise from similarities in uptake mechanisms for different nutrients, whereas 
others may arise from precipitation or immobilization of ions near the plant root or 



274 ACID SOILS 



Table 10.3. Effect of lime and molybedenum on the yield of 
alfalfa hay in the Willamette Valley of Oregon 3 





Average Alfalfa Hay 
(kg ha" 1 ) 


Yield 


Treatment 


Melbourne 
Soil 


Aiken 
Soil #1 


Aiken 
Soil #2 


PKBS* 

PKBS lime (5000 kg ha -1 ) 
PKBS Mo (0.5 kg ha" 1 ) 
PKBS lime + Mo 


1280 
7060 
5980 
7310 


4840 
9030 
8100 
9670 


6010 

9810 

10020 

9790 



" From T. L. Jackson et al. 1 967. Reproduced from Soil Acidity and Liming. 
ASA Monograph No. 12, p. 267, by permission of the American Society 
of Agronomy. 

* P = phosphorus, K = potassium, 13 = boron, and S = sulfur, at recom- 
mended rates for alfalfa in western Oregon. 



within the plant root itself. Chemically similar ions also may compete for absorption 
sites on the plant root surface. 



10.10 MANAGEMENT 

Managing acid soils requires that crop tolerance to soil acidity be weighed against 
the cost of liming. Availability of lime, transportation charges, and the necessity 
and cost of grinding the limestone all influence the quantities that can be applied 
economically. A rule of thumb is to apply sufficient lime initially to raise soil pH to 
the desired range, and then to provide 2 to 5 metric tons lime ha - ' every 3 to 5 years 
to maintain soil pH in that range. Sometimes substituting a more acid-tolerant crop 
may be more economic than liming. 

Soil samples from the surface 20 to 30 cm of soil should be collected every 2 
to 3 years to determine soil pH and to assist with predictions of additional liming 
or fertilization needs. Such sampling is normally done anyway on well-managed 
croplands to maintain adequate levels of available soil P and K. Hence, sampling 
costs should not be assigned entirely to managing soil acidity. Attaining the desired 
soil pH may require 6 to 8 months after lime application and the pH may change 
appreciably for as long as 1 8 months thereafter. Adequate soil water is necessary 
to permit hydroxyl and calcium ion diffusion and to carry out the associated liming 
reactions. 

When the more slowly reacting dolomitic limestone is used, soil pH may increase 
for as much as five years after liming. In general, more finely ground liming materials 
cost more but react faster and more thoroughly with the soil. Finer particles also 
can be dispersed more evenly throughout the soil than can smaller numbers of large 
particles. 



PH AND ION ACTIVITY MEASUREMENTS 



275 



APPENDIX 10.1 PH AND ION ACTIVITY MEASUREMENTS 

The measurement of pH is the most common chemical measurement in soil, biology, 
and aqueous solutions. In addition, electrodes similar in principle and sensitive to 
Na + , K+, Ca 2+ , Mg 2+ , CI - , NOJ , CN _ , F~", S 2_ , and other ions are available. 
Such electrodes are in increasing use and respond roughly to the activity of the ion 
in question. 

In a strict thermodynamic sense, single-ion activities are not measurable. Ad- 
herence to strict thermodynamics, however, can sometimes be unnecessarily lim- 
iting. Ion-sensitive electrodes do respond to changes in the concentrations of ions 
in solution, but they probably do not respond exactly to ion activity. Electrodes can 
also measure spurious potentials under unfavorable conditions. With reasonable care, 
most of these unwanted potentials can be minimized or eliminated. 

The unique property of ion-sensitive electrodes is a membrane between the test 
solution and the electrode sensor that develops an electrical potential, or voltage, in 
response to a change in the concentration of a single ion. The pH electrode, for exam- 
ple, is shown schematically in Fig. 10.5. Other ion-sensitive electrodes differ in the 
composition of the membrane and in the salts necessary to develop the potential. In 
the pH electrode, a silver wire coated with AgCl dips into an HC1 solution. The HC1 
solution is separated from the test solution by a membrane of special glass, usually 
a lithium silicate. Differences in H + activity across this glass membrane cause a dif- 
ference in electrical potential, which can be measured by a sensitive potentiometer. 

The electrode potentials developed by this electrode are the membrane potential 
plus the potential of the Ag — AgCl — HC1 reaction inside the electrode: 



AgCl -He - =Ag + Cr 



(10.9) 



-§= 



Ag wire to 
pH meter 



-Ag CI 

- Standard HCI 

H* - sensitive 
■ glass membrane 






- wire to pH meter 

-Hg-Hg2Cl2 paste 
• Saturated KCI 

Capillary orifice to 
test solution 



FIGURE 10.5. Diagram of the pH and calomel reference electrodes. 



276 ACID SOILS 

This reaction is reversible and, since the activities of Ag and AgCl can be taken as 
unity and the C\~ activity is fixed by the constant HC1 concentration, the Ag — AgCl 
potential (£° = 0.222 V at 25° C) is constant. This potential is accounted for when 
the electrode is standardized against a standard pH buffer solution. 

The electrode develops a second potential across the membrane separating the 
standard HC1 from the test solution. The tiny current flow required by the pH meter 
causes ion exchange at the inner and outer surfaces of the glass membrane and causes 
diffusion of ions across the glass membrane. The electrical current is of the order of 
nano- or picoamperes and diffusion of trace quantities of Na + in the glass apparently 
carries the current. The potential of the pH electrode is 

E = E K( - 0.059 log ° H+ - ' est ( 10. 1 0) 

a H+, std 

where E is the measured potential and is converted to pH units by the scale of the 
pH meter. E K { includes all other (and hopefully constant) potentials which are nul- 
lified by standardizing the system with a standard pH buffer solution. Despite this 
complexity, the potential across the glass membrane can be closely calibrated to the 
approximate value of the H' 1 " activity. 

A second or reference electrode is necessary to complete the electrical circuit. The 
reference electrode is sometimes welded to the pH electrode so that the pair look like 
a single electrode. Reference electrodes are loo often taken for granted; their spurious 
potentials are a common source of error in soil pH measurements. A typical reference 
electrode is also sketched in Fig. 10.5. The wire dipping into the liquid mercury 
makes electrical contact with the pH meter, and current flows from the electrode to 
the solution phase through the reversible reaction (£° = 0.268 V at 25° C): 

Hg 2 Cl 2 + 2e~ = 2Hg 4- 2CF (10.11) 

The Cl~~ activity is fixed by the KC1 concentration (usually saturated KC1). This 
potential is also compensated when the pH electrode system is standardized in a 
standard pH buffer. KC1 diffusion through the orifice makes the electrical contact 
between the reference cell and the test solution. This KC1 connection forms a "salt 
bridge" between the test solution and the reference electrode. 

A junction or diffusion potential always develops when two dissimilar substances 
or solutions of different composition come in contact. An example in this case is 
the saturated KC1 of the reference electrode as it diffuses into the test solution. The 
potential is minimal when the rales of K + and Cl~ diffusion into the test solution 
are equal. Indeed, KC1 was chosen as the reference cell electrolyte because of the 
similarity of K+ and CI - diffusion rates in water. 

When immersed in solution, the reference electrode in Fig. 10.5 usually fulfills its 
role of simply completing the electrical circuit. In a colloidal suspension, however, 
the colloid may cause K + and Cl~ to diffuse at different rales. Because of attraction 
or repulsion by the charged colloid, one ion moves ahead of the other. Ion separation 
at the junction between the electrode solution and the suspension produces a charge 
separation or electrical potential, the liquid-liquid junction potential (£j). Accurate 



BIBLIOGRAPHY 277 

pH measurements require a negligible £j, because such potentials are unpredictable. 
The potentiometer of the pH meter measures all potentials of the circuit and can- 
not distinguish between the H + potential at the glass membrane and the spurious £j 
values. The value of £j at the interface between saturated KC1 and a colloidal sus- 
pension of very low salt concentration can be as high as 240 rnV, equivalent to more 
than 4 pH units. Such extreme £j values are unlikely in soil suspensions, because 
salt concentrations even in highly leached soils are normally at least several mmoles 
per liter. Values of £j greater than 30 mV, an error equivalent to 0.5 pH unit, are 
probably uncommon for soils. Measuring pH in salt solutions of 0.0 1 M or greater 
virtually eliminates E } . 

Another simple way to minimize /?j in soil pH measurements is to allow the tip of 
the reference electrode to contact only the supernatant solution above the colloidal 
phase. The rates of K + and Cl~ diffusion are then unaffected by the colloid. The 
glass electrode, on the other hand, can be placed either in the supernatant solution or 
in the colloidal suspension. The H + activity is the same in both phases, and the glass 
electrode is unaffected by the presence of the colloid. 

The electrodes of a pH measurement circuit can be shown as 

Ag | AgCI | standard HC1 | test solution || saturated KC1 | Hg 2 Cl 2 | Hg 

^membrane ''j 

where each bar represents a phase boundary and where the double bar represents the 
liquid-liquid junction. Other ion-sensitive electrodes differ primarily in the compo- 
sition of the membrane. 

The glass membrane of the pH electrode has proved to be by far the most suc- 
cessful ion-sensitive membrane. The glass has a uniform response to a wide range 
of H + activities (or concentrations), requires little maintenance, is resistant to con- 
tamination, is structurally strong, and is insensitive to interfering ions. The extent of 
interference is denoted by the selectivity ratio, which is the concentration ratio of the 
test ion to interfering ion at which the interfering ion exerts a significant potential at 
the membrane. The selectivity ratio of the pH glass membrane for H + over Li + , the 
most serious interfering ion, is about 10 9 . That is, Li + would cause a significant pH 
error in a pH 9 solution containing 0. 1 M or more Li + . The selectivity ratio for Na + , 
the next most serious interference, is about 10 13 . 

Other ion-sensitive electrodes are not nearly as effective in screening out interfer- 
ing ions. Selectivity ratios are as low as 1, meaning that the electrode is as sensitive 
to the interfering ion as to the test ion. Such measurements are valid only when the 
concentrations of interfering ions are considerably lower than that of the test ion. Ion- 
sensitive membranes are being continually improved and hold considerable promise 
for soil chemical analysis. 



BIBLIOGRAPHY 

Chernov, V. A. 1947. On the Nature of Soil Acidity. Academy of Sciences, Moscow U.S.S.R. 
170 pp. (Translated and published by the Soil Science Society of America, Madison, WI.) 



278 ACID SOILS 

Coleman, N. T., and G. W. Thomas. 1967. The basic chemistry of soil acidity. Agronomy 
12: 1-41 . Excellent review article on acid soil chemistry. 



QUESTIONS AND PROBLEMS 

1. Calculate the relative acidifying tendencies of 100 kg ha -1 of nitrogen as 
(NH 4 ) 2 S0 4 , NH4NO3, and NH 3 . 

2. Based on the data of Table 10. 1 , calculate the times required for Utah bentonite, 
Volclay bentonite, and kaolinite to convert from H + form to >99% Al 3 " 1 " form. 

3. A mineral subsoil of initial pH 4.8 and CEC 76 mmoles(+) kg"" 1 is titrated with 
OH - to pH 6.5. Sketch the variation in CEC with pH that you would expect 
during this process. 

4. For the soil of Problem 3, discuss the probable composition of the exchange 
complex at 

(a) pH4.8 

(b) pH 5.7 

(c) pH6.5 

5. An acidic pesticide of pK 5 is being applied to soils of a given region. Would 
leaching be greatest for soils of 

(a) pH > 5? 

(b) P H6to8? 

(c) pH < 5? 
Why? 

6. A soil of pH 5.5 retains 60 mmol(+) kg of exchangeable bases and has a CEC 
at pH 7 of 80 mmol(+) kg" ' : 

(a) What is its percent base solution? 

(b) What is its approximate CEC at pH 5.5? 

7. A soil has a pH of 5.2, retains 70 mmol(+) kg -1 of exchangeable bases, 10 
mmol(+) kg -1 of exchangeable Al 3+ , and 30 mmol(-) kg -1 of phosphorus at 
pH 5.2, and has CEC and phosphorus-retention capacities at pH 7 of 100(+) and 
15 mmol(— ) kg -1 , respectively: 

(a) What is its percent base saturation? 

(b) What is its percent exchangeable acidity? 

(c) What is its amount of titratable but nonexchangeable acidity? 

8. For the soil of Problem 7, what is the approximate field lime requirement if the 
pH is to be raised to 6.5? 

9. Based on the data of Fig. 1 0.4, how much effect would variation in titration time 
from 0.5 to 48 hours have on the lime requirement of pH 5 Greenfield soil, if a 
final pH of 6.5 were sought and if 5 g of soil were used for the titration shown? 



QUESTIONS AND PROBLEMS 279 

10. Based on the data of Table 10.2, tabulate the approximate pH values required to 
produce 50, 75, and 90% of maximum yield for each of the crops listed. 

11. Calculate the titration curve of a 0.01 M HC1 solution titrated with 0.01 M NaOH 
to verify the strong acid curve of Fig. 1 0. 1 . 

12. Based on Fig. 1 0.4, how are laboratory titration data related to lime requirement 
values in the field? 

13. With a glass electrode H + /Li + selectivity ratio of 10 9 , calculate the pH error if 
1 M LiCl were used to extract exchangeable cations at measured pH values of 6, 
8, and 10. 



11 



SALT-AFFECTED SOILS 



Salt-affected soils are common in arid and semiarid regions, where annual precipi- 
tation is insufficient to meet the evapotranspiration needs of plants. As a result, salts 
are not leached from the soil. Instead, they accumulate in amounts or types detri- 
mental to plant growth. Salt problems are not restricted to arid or semiarid regions, 
however. They can develop in subhumid and humid regions under appropriate condi- 
tions. Basic principles of soil chemistry directly apply to the study and management 
of salt-affected soils. 



1 1 .1 DISTRIBUTION AND ORIGIN 

Salt-affected soils often occur within irrigated lands. In the United Slates, 5 million 
ha of irrigated land are estimated to be salt-affected, mostly in the 1 7 western states. 
A recent survey indicates that as much as one-third of all irrigated lands in the world 
(or approximately 70 million ha) may be plagued by salt problems. When salt prob- 
lems of nonirrigated semiarid and humid regions, greenhouse crops, mine spoils, and 
waste disposal areas are added to these figures, the dimensions of the problem are 
truly impressive. Though one might think that naturally saline areas would be better 
left unfarmed, the typically favorable year-round climates of many such areas, the 
desire to develop all of a farm for crop production, and the expense of installing and 
maintaining a water conveyance system can dictate the reclamation of many saline 
areas. 

The three main natural sources of soil salinity are mineral weathering, atmo- 
spheric precipitation, and fossil salts (those remaining from former marine or la- 
custrine environments). The human activities that add salts to soil include irrigation 
and saline industrial wastes. Seawater encroachment can also harm soils. 

280 



DISTRIBUTION AND ORIGIN 281 

The ultimate source of all soil salts is the exposed rocks and minerals of the earth's 
crust, from which salts have been released during chemical and physical weathering. 
In humid areas, soluble salts are carried down through the soil profile by percolating 
rainwater and ultimately are transported to the ocean or to inland seas. In arid re- 
gions, leaching is generally more localized. Salts tend to accumulate because of the 
relative scarcity of rainfall, high evaporation and plant transpiration rates, or land- 
locked topography. 

Without leaching, in situ weathering of primary minerals would eventually allow 
soluble salts to accumulate to hazardous levels, but this degree of accumulation is 
rare. Salts are released during weathering. Mafic mineral (dark, Mg- and Fe-rich) 
minerals, tor example, are common in arid-region soils. If present in sufficient quan- 
tities, they can increase the salt concentration of slowly percolating waters by as 
much as 3 to 5 mmol(+) L~ ' . In arid regions, the occasional rains that cause the 
weathering are usually sufficient to flush out most of the salts. 

Weathering minerals rarely dissolve congmently (in strict proportion to their com- 
position). Instead, they release their most soluble components first. A mineral high 
in Ca and Mg may therefore initially release significant amounts of Na and K to the 
percolating solution. The water weathering the minerals is usually of sufficient quan- 
tity to carry the soluble salts thus created to the sea, to a landlocked lake, to a nearby 
saline seep, or at least to the average annual depth of wetting of the soil. 

So-called fossil salts can introduce large amounts of salinity into soil and ground 
water. This was dramatized in the 1960s by the Wellton-Mohawk irrigation project 
of Arizona, where saline groundwaters were discharged into the Gila River after ir- 
rigation raised the groundwater level in a valley underlain by saline deposits. The 
drainage water mixed with the Colorado River and significantly increased the river's 
salinity. Downstream fanners in Mexicali, Mexico, were understandably angered 
when the more saline water damaged their irrigated crops. Fossil salts dissolving in 
percolating waters contribute materially to the salinity added to the Colorado River 
from several irrigation projects along its upper reaches. 

Fossil salts can also be dissolved when water-storage or water-transmission struc- 
tures are placed over saline sediments. The Lake Mead reservoir behind Hoover Dam 
in southern Nevada overlies deposits of gypsiferous sediments. Dissolution of this 
gypsum substantially increases the salinity of the Colorado River during its passage 
through the reservoir. 

Appreciable salt can also be deposited in some areas from the atmosphere. Rain 
droplets form around tiny condensation nuclei such as salt or dust particles. The total 
salt concentration of rainfall may be as high as 50 to 200 mg L - ' near the seacoast, 
but rapidly decreases to only a few mg L~' in the continental interior. The exact 
pattern of the decrease depends on local topography and weather patterns. Changes 
in composition of the rainfall also occur. The salts in rain near the seacoast are high 
in Na, CI, and Mg. Inland precipitation is dominated by Ca and Mg sulfates and 
bicarbonates. 

The quantities of salt added from the atmosphere to arid and semiarid regions 
may amount to only a few kilograms per hektar per year, but the amounts introduced 
over periods of tens to thousands of years can be substantial. The vegetation of such 



282 SALT-AFFECTED SOILS 

areas normally has reached a balance with incoming precipitation, so salts tend to 
accumulate below the surface at the average depth of soil wetting. They can then 
be flushed from the soil at relatively high concentrations when a period of particu- 
larly high rainfall occurs, or when human activities tend to change the annual water 
balance. Such changes have contributed to the saline seeps now common in cer- 
tain controlled-brush and overgrazed portions of Australia, and in summer-fallowed 
wheat areas of eastern Montana and the Dakotas. The conditions causing saline seeps 
and saltpans on an Australian landscape are shown in Fig. 11.1. Salt accumulates at 
the seep on the slope because of arrested soil drainage. The rates of salt input and 
evapotranspiration at the seeps are greater than the rales of leaching or down slope 
runoff. 

The saltpan in the basin of Fig. 11.1 exemplifies a more common occurrence of 
soil salinity. Soils in low-lying areas, even in arid regions, may have high water 
tables. Water from groundwater tables within a few meters of the surface can move 
by capillarity to the soil surface, where it evaporates and leaves behind its salts. 
Figure 1 1.2 shows an example of salt distribution above a water table 90 cm below 
the soil surface. The soil salinity concentration is expressed as electrical conductivity, 
the common method of measurement. 

Large-scale examples of water collection and evaporation are the Great Salt Lake 
of Utah, a remnant of ancient landlocked Lake Bonneville that once covered much 
of the western United States; the Caspian Sea in Asia; Lake Chad in Africa; and 
Lake Ayre in Australia. More common examples are the fringes of salt accumulation 
along arid-region rivers and drainageways and small playas in many arid regions. Salt 
accumulation on a local level also includes "slick spot" patches of sodic (sodium- 
rich) soil, which can be accompanied by marked soil morphological changes as a 
result of repeated clay swelling and migration. 

Many present-day salt-affected soils result from human activities. Salts commonly 
are transported from areas of overirrigation to accumulate in poorly drained areas. As 
drainage waters or irrigation return flows (drainage waters) evaporate, high concen- 
trations of salts may remain. An example is the Salton Sea of southern California, 
which formed initially after a break in the dikes of the Colorado River during the 
early 1900s. It has become highly saline during subsequent accumulation and evap- 
oration of irrigation return flows from the nearby Coachella and Imperial Valleys. 
Salts also accumulate in underirrigated fields, particularly if relatively saline irriga- 
tion waters are used. The salt concentration of the soil solution increases steadily as 
water is removed during plant growth. Proper irrigation management includes peri- 
odic irrigation with water in excess of plant needs, to leach accumulated salts from 
the plant root zone. Since water in these regions is scarce and expensive, minimizing 
the amount of this leach water is important. 

Such human activities as oil-field development, waste-spreading operations, and 
fertilization can also add sizeable quantities of soluble salts to soils. Development of 
tidal or formerly marine areas can lead to salinity from saltwater intrusion whenever 
freshwater is insufficient to keep out seawater. The seaward flow of freshwater is 
often decreased by pumping from wells and diversion of streams for irrigation. In 
the Netherlands, treated municipal wastewater is pumped into the ground to prevent 




c 
o 

' V. 

CO 



CO 



T3 

c 
to 
« 

0) 
CD 



3 

<d 
o 

To 



D> 
O 
O 



O 

E 
o 



% 




o 





-C 


o 






CO 


7 




ti 




00 


en 




_> 




< 










u: 


m 






o 


5 


w 


<TI 


~j 


V 




CD 

m 


S 
^ 


<i> 




x: 


■si- 




ll) 


c 


O) 






<l> 




o 


CO 


■>. 




o 


U> 


o 


i& 






o 


<B 


o 


XI 




<n 




CD 


I 


UI 


^1 


e 


7- 


p 


T~ 


'•— 


Ui 


x> 


3 


"5. 


o ts 


IL 


< 



283 



284 



SALT-AFFECTED SOILS 



EC. , dS/m 
K> 20 30 40 50 60 70 




FIGURE 11.2. Typical salinity profile in soil exposed to a high water table. (From R. S. Ayers 
and D. W. Westcot. 1976. Water Quality for Agriculture. Food and Agriculture Organization of 
the UN, Rome.) 



seawater intrusion along the coast. The soil pores prevent any mixing of the freshwa- 
ter with saltwater so the freshwater is an effective dam against seawater movement 
inland. 



1 1 .2 IRRIGATION WATER QUALITY 

The major cause of soil salinizalion is unsatisfactory irrigation and drainage. Various 
systems have been proposed to classify the quality of irrigation and drainage wa- 
ters. Irrigation involves applying water to the soil surface, displacing unused water 
downward through the soil during subsequent irrigations, and eventual emergence of 
drainage waters from bottom of the plant root zone. Some water is lost during evap- 
oration at the soil surface, and the plant removes considerably more water during 
transpiration. Although plants absorb some salts, both evaporation and transpiration 
increase the residual concentration of dissolved salts, so the salt concentration of the 
soil solution increases with soil depth. 

Typical salt distributions in irrigated soil profiles are shown in Fig. 1 1.3. As the 
proportion of irrigated water passing through the root zone (the leaching fraction) 
increases, so does the depth of soil that has essentially the same salt concentration as 
the irrigation water. As the leaching fraction increases, salt accumulation is pushed 
down to lower depths. 

As the salt concentration of the soil increases, so does the potential for salinity 
effects on plant growth. Early appraisals of the salinity of irrigation waters were 
in terms of total dissolved solids (TDS). The TDS were determined by evaporat- 
ing a known volume of water to dryness. The presence of hygroscopic water in the 
resultant salt mixtures made the values for TDS strongly dependent on the drying 
conditions. The concentration of salts in most irrigation waters is less than 1000 mg 



IRRIGATION WATER QUALITY 



285 



0.5 



I 

y- 

0- 



EC, dSfm 
10 15 



20 



n is! • *~ 


i 


\ |N 




\ \\ 




\\ i ^ 




\\ v \ 




\ \ \ ^ 




\ \\ -- 




\ \ 




\ \^ ^*"^^ 




\ V"--^ ^-v.o.10 


\ N 


N. 


\ \ 




0.13\ \ 


\ \ 


\ \ 
\ \ 


0.06\ \ 


\ \ 


\ 1 


\ \ 


\ 1 


\ \o.25 


I 1 
1 


I \ 


1 1 



1.0 



FIGURE 1 1 .3. The steady-state profile oi soil salinity, expressed as the electrical conductivity 
of the saturation extract, in lysimeters. The irrigation water has EC values of 2 (solid lines) and 
4 (dashed lines) dS m~ 1 . Numbers on the figure are the respective leaching fractions. 



L -1 TDS. Over half of the waters used for irrigation in the western United States 
have TDS values less than 500 mg L~' ; less than 10% have values greater than 1500 
mg L~ ' . Groundwaters used for irrigation are usually higher in TDS than surface 
waters. Some groundwaters have been used successfully for irrigation despite TDS 
values approaching 5000 mg L~ ' . 

More recently salinity has been measured in terms of the electrical conductiv- 
ity (EC) of a solution. In addition to overcoming some of the ambiguities of TDS 
measurements, the EC measurement is quicker and sufficiently accurate for most 
purposes. To determine the EC, the solution is placed between two electrodes of 
constant geometry, including constant distance of separation. When an electrical po- 
tential is imposed, the electrical current varies directly with the total concentration 
of dissolved salts. The current is inversely proportional to the solution's resistance 
and can be measured with a resistance bridge. Conductance is the reciprocal of re- 
sistance and has units of reciprocal ohms or Siemens (formerly mhos). The EC of the 
saturation extract of the soil measures the salinity of the soil. 

The measured conductance is a result of the solution's salt concentration and the 
electrode geometry in the measuring cell. The effects of electrode geometry are em- 
bodied in the cell constant, which is related to the distance between electrodes and 
their cross-sectional area. The cell constant is measured by calibration with KC1 so- 
lutions of known concentration. The conductivity of KC1 solutions is available in 
published tables. For example, calibration might yield a cell constant of 2.0 cm -1 . 
A test solution that measures 2000 Q. resistance (conductance of 1/2000 fir 1 or 
0.0005 Siemens) in this cell has a conductivity of 1 .0 dS m~ ' , or 1.0 mmho cm - ' . 

The former unit of electrical conductivity was millimhos per centimeter (mmho 
cm -1 ). In SI units, the unit of conductivity is Siemens (1 S = 1 mho, so that 1 dS 



m 



-i _ 



= 1 mmho cm ). When dealing with rainwater or with river water of low 



286 SALT-AFFECTED SOILS 

salinity, results were also reported as micromhos per centimeter (fimho cm -1 ). A 
water with an EC of 0.2 mmho cm - ' has an EC of 200 yxmho cm" ' or 0.2 dS m~ ' . 

Now, in situ soil EC measurements are being made by (1) sensors embedded in 
porous ceramic, thus maintaining solution contact with the electrodes; (2) groups of 
electrodes (commonly four) placed across the soil surface to measure the salinity of 
underlying soils; or (3) mobile electromagnetic "wands," which can be carried across 
the landscape to give similar information. 

Several empirical relationships have been developed for converting one type of 
water quality analysis to another. For solutions in the EC range from 0.1 to 5 dS 
m-'. 

Sum of cations or anions (mmol(+ or -) L" 1 ) % EC (dS m~') x 10 (11.1) 

and 

TDS (mg L" 1 ) s* (EC (dS m -1 ) x 640 (l 1.2) 

For soil extracts in the EC range from 3 to 30 dS m - ' , 

OP (bars) » EC (dS m -1 ) x (-0.36) (11.3) 

where OP is the osmotic potential, or the negative of the osmotic pressure of the 
water. The osmotic pressure or osmotic potential most directly measures the effects 
of salinity on plant growth. An irrigation water containing 3 mmol(-f) L _1 Ca 2+ , 2 
mmol(-f) L _1 Mg 2+ , and 3 mmol(+) L~' Na + has 8 mmol(+) L" ' total cations, an 
EC of approximately 0.8 dS m -1 , a TDS value of approximately 510 mg L _1 , and 
an OP of approximately —0.3 bars or -30 kPa. 

Whenever complete chemical analyses are provided for soil extracts or irrigation 
waters, the sum of major cations (mmol(+) L _1 ) should approximately equal the 
sum of all major anions (mmoI(— ) L _1 ). Repeated exact agreement, however, in- 
dicates that one ion is being determined by difference. This is usually sulfate for 
recent analyses, or sodium for older analyses. Also, reported concentrations of car- 
bonate should be negligible at solution pH >9. In the water-supply literature, Ca 
plus Mg concentrations are reported as hardness, the chemically equivalent quantity 
of CaC03 in milligrams per liter. Concentrations of bicarbonate plus carbonate may 
be reported as alkalinity, the equivalent acid-neutralizing capacity of the water. 

11.2.1 Sodium Hazard 

Another important measurement of water quality is its relative amount of sodium 
(sodicity). Irrigation waters with a high sodium content tend to produce soils with 
high exchangeable sodium levels. Such soils crust badly and swell and disperse, 
greatly decreasing the soil's hydraulic conductivity, or water permeability. Clay par- 
ticles disperse and plug the soil-water flow channels, as does swelling of clay parti- 
cles. Decreased permeability interferes with the drainage required for salinity control 
and with the water supply and aeration required for plant growth. 



IRRIGATION WATER QUALITY 287 

Early estimates of sodicity were based on sodium content. Because of the strong 
preference of most soil particles for divalent cations over monovalent cations, how- 
ever, waters with high Na contents may still produce relatively low exchangeable Na 
levels in soils, if the Ca + Mg concentration is appreciable. 

Cation exchange equations contain the ratio of the monovalent cation concentra- 
tions to the square root of the divalent cation concentration, or the square of this 
ratio. The equation may involve ion activities rather than concentrations, and may 
include corrections for ion pairs. For most field practice, however, the ratios of total 
ion concentrations alone are sufficient. 

Workers at the U.S. Salinity Laboratory proposed the sodium adsorption ratio 
(SAR) to characterize the sodium status of irrigation waters and soil solutions: 

SAR = — ^^^ (11.4) 

[[Ca 2 ++Mg 2 +]/2]' /2 

where the brackets indicate that the concentrations are in millimoles(+) per liter. The 
Ca plus Mg term is divided by two because most ion-exchange equations express 
concentrations as moles per liter or mmoles per liter, rather than as millimoles(+) 
per liter. Combining Ca and Mg is not strictly correct but seems to cause little loss 
of accuracy. The combination is necessary because many early water analyses com- 
bined Ca with Mg, and it is justified because these two divalent cations behave sim- 
ilarly during cation exchange. Recent work has distinguished between the "true" 
SAR (involving ion activities), and the "practical" SAR (SARp) involving the ratio 
of concentrations. 

The exchangeable sodium status of soils can be predicted quite well from the SAR 
and a Gapon-type exchange equation: 



[NaX] _ tfotNa-' 

[CaX 4- MgX] ~ [ lCa 2+ + Mg 2 +]/2] ] 



ESR = TT^T^ = TTZ^T^-^Tp. = KgSAR (1 1.5) 



where ESR is the exchangeable sodium ratio of the soil, X is the soil, the exchange- 
able ion concentrations are in milIiinoles(+) per kilogram, and Kg is the Gapon 
exchange constant. The range of Kq is commonly 0.0 10 to 0.015 (L mmol)~ ' ^ . The 
values of ESR of the soils and ESP/100 (exchangeable sodium percentage) of the ir- 
rigation water are approximately equal for many irrigated soils at ESP values below 
25 or 30% . The exact relation between the two parameters is 

„ 100 ESR 
ESP=- — — (11.6) 

1 -+- ESR 

Water having an EC of I dS m -1 and a sodium percentage ([mmol(+) L -1 of 
Na + ]/lmmol(-|-) L _1 of total cations] x 100) of 92% has an SAR of approximately 
15. 



288 



SALT-AFFECTED SOILS 



11.2.2 Bicarbonate Hazard 

Another property related to the sodium hazard of irrigation waters is the bicarbonate 
concentration. Bicarbonate toxicities associated with some waters generally arise 
from deficiencies of iron or other micronutrients caused by the resultant high pH. 
Precipitation of calcium carbonate from such waters, 



.,2+ 



Ca z+ + 2HCO" = CaC0 3 + H 2 + C0 2 



(11.7) 



lowers the concentration of dissolved Ca, increases the SAR, and increases the 
exchangeable-sodium level of the soil. The CaCO.-? precipitation can be accounted 
for by the adjusted SAR: 



Adjusted SAR = SAR x [1 + (8.4 - pH c )l 



(11.8) 



where pH c is as defined and discussed in Appendix 11,1. The concept of an adjusted 
SAR has found widespread applicability. Figure 1 1 .4 shows the relation of the ad- 
justed SAR of the irrigation water to the SAR of the saturation extracts for a group 
of Middle Eastern soils after three years of irrigation. 

Early workers used the residual sodium carbonate (RSC) to predict the tendency 
of calcium carbonate to precipitate from high-bicarbonate waters and thus create a 
sodium hazard. The RSC was defined as 

RSC - IHCO3 + CO^-] - [Ca 2+ + Mg 2+ ] ( 1 1 .9) 

with all values in millimoles(±) per liter. Waters of RSC greater than 2.5 were con- 
sidered hazardous under all conditions. RSC values between 1.25 and 2.50 were 



40 


- 



















35 












O 






d 30 

O 

(/) 

O 25 


















Ul 












x^ 






1- 


















< 


















<r 20 


^ 

















z> 












^0 






Si 








O 










<r> 15 


- 

















O 










O 








or 10 


- 




^S^O 












< 


















</} 




O ^S' 














5 

n 




1 


1 


1 


1 


1 1 1 







10 15 20 25 

ADJ. SAR OF IRRIGATION WATER 



30 



35 



40 



FIGURE 11.4. Influences of adjusted SAR of the irrigation water upon the SAR of saturation 
extracts from Pakistani soils after three years of cropping. (From R. S. Ayers and D. W. Westcot. 
1976. Water Quality for Agriculture. Food and Agriculture Organization of the UN, Rome.) 



CHARACTERIZING SALT-AFFECTED SOILS 289 

considered potentially hazardous, and waters with RSC values less than 1 .25 were 
considered safe. These predictions worked reasonably well. 

The main disadvantage of the RSC was that it treated all bicarbonate in the wa- 
ter as if it would precipitate. This was incorrect, for the amount of bicarbonate that 
precipitates depends on the degree to which salts are concentrated by evapotranspi- 
ration in the plant root zone. As an extreme example, if no water evapotranspired, 
all the bicarbonate would pass through the soil unchanged. Conversely, if all of the 
water evapotranspired, all of the bicarbonate would precipitate. Hence, the quantity 
of bicarbonate precipitating depends on the proportion of water percolating through 
the soil, or the leaching fraction. 

1 1 .2.3 Other Toxic Solutes 

Irrigation waters also contain potentially toxic ions such as boron, lithium, sodium, 
and chloride. The boron concentrations of irrigation waters are particularly impor- 
tant, because many crops are susceptible to even extremely low concentrations of this 
element. The differences between deficient and toxic B concentrations are only a few 
milligrams per liter. Sodium and chloride ions also are hazardous to fruit and berry 
crops and to other woody plants. Their ranges of hazardous concentration are consid- 
erably higher than for boron. In addition to absorption through roots, toxic ions also 
can be taken up by foliage. Sprinkling water high in sodium or chloride on the leaves 
of horticultural plants and vegetables, fruits, and berry crops can cause plant damage 
as the water evaporates. Although Li is potentially a problem in the Coachella Valley 
of California, the management that controls B, Na, and CI also prevents Li toxicity. 



11.3 CHARACTERIZING SALT-AFFECTED SOILS 

The sodium status of soils is generally best described by the soil's exchangeable 
sodium percentage. Measuring ESP, however, is tedious and subject to error. The 
concentration of "soluble" sodium in the bulk solution must be measured and sub- 
tracted from the total quantity of sodium extracted to obtain the exchangeable 
sodium. Soluble sodium can be measured in the saturation extract, but anion exclu- 
sion can produce excessive soluble sodium concentrations in extracts from high-clay 
soils. This results in low ESP values. Incomplete removal of the index salt solution 
during the wash step of CEC determinations can lead to high CEC values and there- 
fore to low ESP estimates. Hydrolysis during removal of the index salt solution, 
trapping of NH* from the index solution between soil mineral lattices, and calcium 
carbonate or gypsum dissolution in the index or replacement solutions can all lead 
to low CEC values and hence to high ESP estimates. 

Still another special problem in CEC and ESP determinations occurs for soils 
of high pH containing significant amounts of the slightly soluble zeolite minerals. 
Zeolites such as analcime and leucine contain replaceable monovalent cations in their 
crystal lattices. These structural cations are readily displaced by other monovalent 
cations, but not by divalent cations. If a monovalent cation is used as the index or 



290 SALT-AFFECTED SOILS 

replacement cation, the amounts of sodium or ammonium extracted are erroneously 
high, for many of the extracted cations would not be available for normal exchange 
by divalent or trivalent ions. This problem should be suspected whenever soils of high 
pH have unusually high ESP-SAR relationships. In such cases, estimating the true 
ESP from the SAR of the saturation extract may be more accurate than measuring 
the ESP directly. 

As a result of these potential errors in soil ESP determinations, and because of the 
generally good relationship between SAR of the soil solution and ESP of the soil, the 
SAR of the saturation extract is normally a satisfactory index to the exchangeable- 
sodium status of salt-affected soils. Since the saturation extract is already required 
to determine the EC, using the SAR requires only that a few additional chemical 
determinations be made on this extract. In fact, when the quantity of solution or the 
cost of analyses is limiting, the SAR can be estimated from the EC and either the 
sodium or the calcium plus magnesium concentration alone. The EC reflects the to- 
tal cation concentration, and saline-soil extracts typically contain few cations other 
than sodium, calcium, and magnesium. A solution having an EC of 0.6 dS m _1 and 
a sodium concentration of 3 mmol(+) L~ ' would have a total salt concentration of 
approximately 6 mmol(+) L~ ' , a calcium plus magnesium concentration of approx- 
imately 3 mmol(-f) L _1 , and an SAR of approximately 3/(3/2) l/2 = 2.5. 

The traditional classification of salt-affected soils in the United States has been 
based on the soluble salt (EC) concentrations of extracted soil solutions and on the 
exchangeable sodium percentage of the associated soil. The dividing line between 
saline and nonsaline soils was established at 4 dS in -1 for water extracts from satu- 
rated soil pastes. Salt-sensitive plants, however, can be affected in soil whose satu- 
ration extracts have ECs of 2 to 4 dS m _l . The Terminology Committee of the Soil 
Science Society of America has recommended lowering the boundary between saline 
and nonsaline soils to 2 dS m _1 in the saturation extract. 

The traditional and recently proposed classification categories for salt-affected 
soils are given in Table 11.1. Saline (white alkali) soils are those in which plant 
growth is reduced by excess soluble salts. These soils can be converted to normal 
soils by leaching the excess salts from the plant root zone. The pH of saline soils 



Tabfe 11.1. Traditional and proposed classifications of salt-affected soils* 





Normal 


Saline 


Sodic 


Saline-Sodic 




Soiis 


Soils 


Soiis 


Soiis 


Traditional 


EC<4dSirr' 


EC>4dSm _1 


ESP> 1 5% 


EC>4dSm-' 


classification 


ESP < 1 5% 






ESP> 15% 


Proposed 


EC <2dSm _l 


EC>2dSm _l 


SAR> 15 


EC>2dSm _l 


classification 


SAR < 1 5 






SAR > 15 



"From Terminology Committee. 1973. Glossary of Soil Science Terms, Soil Science Society of America, 
Madison, WI. 



EFFECTS OF SALTS ON SOILS AND PLANTS 291 

generally is less than 8.5, and they are normally well flocculated (i.e., as permeable as 
might be expected from soil texture alone). Plants growing on such soils may appear 
stunted and have thickened leaves and a dark green color. Substantial reductions in 
plant growth can occur without appreciable changes in plant appearance. 

Soils containing both high soluble-salt and high exchangeable-sodium levels 
are called saline-xodic. Such soils also reduce plant growth because of their high 
soluble-salt content. Because the soluble salts prevent hydrolysis, the pH of saline- 
sodic soils is typically less than 8.5. The main hazard occurs when these soils are 
leached to remove salts. Leaching removes the salts faster than it removes exchange- 
able Na, causing conversion to sodic soils. This can severely reduce soil permeability 
or hydraulic conductivity and affect plant-water relations and the ability to leach for 
salinity control. 

Sodic (black alkali) soils are a particularly difficult management problem. The 
water permeability of these soils to water is very slow. The pH of sodic soils is 
commonly greater than 9 or 9.5, and the clay and organic fractions are dispersed. 
Dispersed organic matter accumulates at the surface of poorly drained areas as water 
evaporates and imparts a black color to the surface, hence the name "black alkali." 
Sodic soils are found in many parts of the western United Slates. In some locations 
they occur in small patches, "slick spots," less than 0.5 ha in extent. Such patches 
occupy slight depressions, which become accentuated as surface soil particles dis- 
perse and are blown away by wind erosion. The percolation of insufficient water to 
satisfy plants and to control salinity is the main problem associated with sodic soils. 
In addition, their relatively low soluble-salt concentrations and high pH values can 
result in direct Na toxicities to the most sensitive plants. 



1 1 .4 EFFECTS OF SALTS ON SOILS AND PLANTS 

The main effect of soluble salts on plants is osmotic — plants must expend large 
amounts of energy to absorb water from the soil solution, energy that would oth- 
erwise be used for plant growth and crop yield. The plant root contains a semiperme- 
able membrane permitting water to pass but rejecting most of the salt. Thus, water is 
osmotically more difficult to extract from increasingly saline solutions. Plants grow- 
ing on saline media can somewhat increase their internal osmotic concentrations by 
producing organic acids or by absorbing salts. This process is called osmotic adjust- 
ment. The effect of salinity on the plant appears primarily to be energy diversion 
from growth processes to maintain the osmotic differential between the interior of 
the plant and the soil solution. One of the first processes from which this energy is 
diverted is cell elongation. Leaf tissue cells continue to divide but do not elongate. 
The occurrence of more cells per unit leaf area accounts for the typically dark green 
color of osmotically stressed plants. 

The relative growth of plants in the presence of salinity has been termed their 
salt tolerance. Earlier data were summarized by separating plants into several salt- 
tolerant groups. Subsequent listings were in terms of relative plant growth at various 
salinity levels (EC) of the soil's saturation extract (Table 1 1.2). Some recent listings 



292 SALT-AFFECTED SOILS 

Table 11.2. Salt tolerance of plants 3 



EC (dS m -1 at 25°C) at Which 
Yield Will Be Decreased by* 



Crop 



10% 



25% 



50% 



Forage Crops 

Bermudagrass^ (Cynodon dactylon (L.) Pers.) 

Tall wheatgrass (Agropyron elongation (Host) Beauv.) 

Crested wheatgrass (Agropyron desertonwi (Fisch. ex Link) Schult.) 

Tall fescue (Festuca arundinacea Schreb) 

Bailey, hay rf (Hordeum vulgare L.) 

Perennial ryegrass (Lolium perenne L.) 

Harding grass (Pfmlaris stenoptera Hack) 

Narrow-leaf birdsfoot trefoil (Lotus tenuifolius (L.) Reich) 

Beardless wild rye (Elymus triticoides Buckley) 

Alfalfa (Medicago sativa L.) 

Orchardgrass (Dactylis glome rata L.) 

Meadow foxtail (Alopecurus pratensis L.) 

Alsike and red clovers (Trifolium hybridum L. and T. pratense L.) 

Field Crops 

Barley, grain'' (Hordeum vulgare L.) 
Sugarbeet 6 (Hordeum vulgare L.) 
Cotton (Gossypium hirsutum L.) 
Safflower (Carthamus tinclorlus L.) 
Wheal (Triticum aeslivum L.) 
Sorghum (Sorghum vulgare Pers.) 
Soybean (Glycine max (L.) Merr.) 
Sesbania (Sesbania macrvcarpa Muhl.) 
Sugarcane (Saccharum officinarum L.) 
Rice, paddy'' (Oryza saliva L.) 
Corn (Zea mays L.) 
Broadbean (Viciafaba L.) 
Flax (Linum usitatiisimum L.) 
Field bean (Phaseolus vulgaris L.) 



13 


16 


18 


11 


15 


18 


6 


11 


18 


7 


10.5 


14.5 


8 


11 


13.5 


8 


10 


13 


8 


JO 


13 


6 


8 


10 


4 


7 


11 


3 


5 


8 


2.5 


4.5 


8 


2 


3.5 


6.5 


2 


2.5 


4 


12 


16 


18 


10 


13 


16 


10 


12 


16 


8 


n 


12 


7 


10 


14 


6 


9 


12 


5.5 


7 


9 


4 


5.5 


9 


3 


5 


8.5 


5 


6 


8 


5 


6 


7 


3.5 


4.5 


6.5 


3 


4.5 


6.5 


1.5 


2 


3 




(i 


continued) 



have been given instead in terms of EC at the point of initial yield decline and in 
terms of percent yield decrease per unit increase in salinity beyond this threshold. 
Most yield data were obtained from uniformly salinized field plots having nearly 
constant salinity with depth. Actual distributions under field conditions more closely 
resemble those from the plots in Fig. 1 1.3, where the plant can extract most of its 
water from the least-salinized portion of the profile. 

As evidenced by the footnotes to Table 1 1 .2, some plants are particularly sensi- 
tive to salinity during the germination or seedling stages when a restricted root zone 
makes the plant extremely vulnerable to osmotic stress. Seedbed shape is often mod- 
ified for such crops to minimize salt accumulation in the vicinity of young seedlings 
(Figure 1 1 .5). Alternate-furrow irrigation (where only one side of the crop row is 
irrigated at any one time) can also be used to flush salts past the young seedling if 



EFFECTS OF SALTS ON SOILS AND PLANTS 



293 



Table 11.2. (Continued) 



EC (dS m -1 at 25°C) at Which 
Yield Will Be Decreased by* 



Crop 



10% 


25% 


50% 


Vegetable Crops 






8 


10 


12 


5.5 


7 


8 


4 


6.5 


8 


4 


fi 


8 


1 2.5 


4 


7 


2.5 


4 


6 


2.5 


4 


6 


2.5 


3.5 


6 


2 


3 


5 


2 


3 


5 


2 


3.5 


4 


1.5 


2.5 


4 


1.5 


2 


3.5 



Beets c ' (Beta vulgaris L.) 

Spinach (Spinacia oleracea L.) 

Tomato (Lycopersium esculentum Mill.) 

Broccoli (Brassica oleracea var. italica L.) 

Cabbage (Brassica oleracea var. capitata L.) 

Potato (Solatium tuberosum L.) 

Sweet corn (Zea Miayj L.) 

Sweet potato (Ipomoea batatas (L.) Lam.) 

Lettuce (Lactuca saliva L.) 

Bell pepper (Capsicum annum L.) 

Onion (Allium cepa L.) 

Carrot (Dancus carota L.) 

Green bean (Plmseolus vulgaris L.) 



"From L. Bernstein. 1964. Sa/r Tolerance of Plants. U.S. Department of Agriculture Information Bulletin 

283. 

*In gypsiferous soils, EC values causing equivalent yield reductions will be about 2 dS m _l greater. 

^ Average for different varieties. Suwannc and Coastal bennudagrass are about 20% more tolerant, and 

Common and Greenfield are about 20% less tolerant, than the average. For most crops, varietal differences 

are relatively insignificant. 

^Less tolerant during the seeding stage. Salinity at this stage should not exceed an EC of 4 to 5 dS m - ' . 

''Sensitive during germination, when salinity should not exceed 3 dS m" 1 . 



SOIL SALINITY AT PLANTING TIME 
(dsm-') 

8 



16 



SINGLE 
ROW 8ED 



DOUBLE 
ROW BEO 



jL- Ju 




Seeds germinate 
^Salt accumulation 




FIGURE 11.5. Effects of bed shape on seedling emergence at various salinity levels. (L. Bern- 
stein et al. 1955. U.S. Dept. Agric. Bull. ARS-41-4, 16 pp.) 



294 SALT-AFFECTED SOILS 

single rows are used. If double-row beds are used, alternate-furrow irrigation can 
flush salts to the vicinity of the bed edge opposite the irrigated furrow and hence 
stress the seedlings near this edge. Drip irrigation, though generally flushing salts 
to the periphery of the wetted soil volume, can also lead to serious salinity prob- 
lems when high rates of fertilizer are added through the drip lines, upon replanting, 
or whenever rainfall flushes accumulated salts toward previously nonstressed plant 
roots. 

In addition to the general osmotic effects summarized in Table 1 1.2, many plants 
are sensitive to specific ions in irrigation waters or soil solutions. Boron toxicity is 
probably the most common. Table 11.3 lists some plants according to their sensitivity 
to the B concentration of irrigation water. Boron is more difficult to control than is 
salinity in general because it leaches more slowly than more soluble salts. 

Direct sensitivity to exchangeable or soluble sodium is more apparent at low salt 
levels, and therefore is difficult to differentiate from the effects of sodium on soil per- 
meability. For plants that are extremely sensitive to sodium, as little as 5% exchange- 
able sodium may lead to toxic accumulations of sodium in leaf tissues (Table 11. 4). 

Chloride toxicity appears to be similar to Na toxicity. Excessive accumulations 
in tissues near plant tips, the end of the plant transpiration stream, lead to necrosis, 
death of leaf tips and margins, and eventual death of the plant. Some plants are able to 
screen out such ions through their root membranes. In addition, different rootstocks 
may possess varying abilities to exclude sodium or chloride from above-ground parts 
(Table 1 1 .5). Some grape rootstocks exhibit up to 30-fold differences in their abilities 
to exclude chloride ions. Selection of a root stock that screens out ions may prevent 
toxic accumulations in plant tops. 

A third mechanism for salt injury to plants is nutritional imbalances. An example 
is the bicarbonate toxicities reported for some saline environments. These result pri- 
marily from reduced Fe availability at the high pH common in high-bicarbonate soils, 
rather than from the bicarbonate ions themselves. The nutritional needs of plants may 
also vary with the types of salts present. For example, high Na levels could lead to Ca 
and Mg deficiencies. The high pH levels of sodic soils can accentuate deficiencies of 
many of the microelements. High soil pH levels also might lead to high concentra- 
tions of soluble aluminium, such as the aluminate (Al(OH)^") species. Salt tolerance 
also can vary with soil fertility, and especially when inadequate fertility limits yields. 
Nutritional effects of salinity on plants are poorly understood at present, however. 
Many of the supposed consequences are still largely speculative. 



1 1 .5 SALT BALANCE AND THE LEACHING REQUIREMENT 

Management of salt-affected soils once centered around maintaining the salt bal- 
ance. This concept dictates that the quantity of salt leaving an area be equal to, or 
greater than, the quantity of salt entering the area. The concern was justified by the 
difficulty in maintaining long-term agriculture for many irrigated areas of the world, 
such as the Tigris and Euphrates valleys of Iraq, where farming has taken place for 
several millenia. Some irrigation projects, however, appear able to operate indefi- 






to 

£ 

TO 

c 
o 



c 

ID 
O 

c 
o 
o 



o 

X! 



w 

c 
in 
c 

TO 

n. 
•5 

0) 

o 

c 

TO 

1. 

c> 

c 

2 
o 

00 



.o 



■3 

I 

CO 



0Q 

*5> 



1 






g 



f2 



a. 
a. 



J3 S? 

CA >Q 

~ 3 

OB *» 

C ^ 

™ -s 

o c 

c -S 

.2 ^ 

5 a. <u 

*> s _c 

a. « o 

S> H 'S 

o to a 

C 

a 



TO 



ill 



13 

a c 

H s 

■5 3 
? s 

§3 

£ S 

Is ° 

to «3 

b E 
2 < 



a. 
a. 



1 


f-^ 


•a 
c 


c 


-51 


TO 





"3 


a 


S 


g 





t*J 


S 


^ 


3 


TO 


3 




-fs 


3 


^ 


ftj. 


S 


a. 







2 "3 



a. > 



a. 



!3 3 

5 «. 



a a, < O 



2 I 

o -5 
■a 



Sȣ Oh u a. 



to 



5 .3 

■s I 



w o 

OB 03 
co O 

63 



I 
I 

L- 



a, 
a* 



3 

3 
c 



fc 



.£ S 



< a. 



3 

c _ 

-5! ^? 

3 3 

« S 5" 

* -S 2 
3 c a 

" ^ 3 

X .5? C 

a. % a 



CD 



2 & j= 

03 H> <A 

O £ « 

H M o! 



1 § I 

■a c g 

w o <u 

c^TB — 

■a m «j 

« TO°^ 

[i. O! O 



~ii 


60 




a 




21 






Co 




> 


3 










s 

3 




2 
S 


1 

a 


■a 


■ii 


g 


3 


1 


£ 




52- 


e 




TO 


I 






2. 

a 



c 

~ 8- 

« S 

00 -~ 

■s ^ 
S 

^S a. 
« E 



TO t— OS TO .3 3 

« ^ U 5 O N (E 



J" 1? -— ■ 

S2S 

? a s 

% *» s 
3 c -5 

I S <u 

II § 

&8. i g 

C< 1U TO 

u IS .S 

CO C/D J 






s£ .S "S --n 





/-n 


o 1 


s; 


<j 


i^ 


a 


s 


.2 
1 


01 

3 
00 

2 


a 
c 


.a 


a 
^0 
3 
> 


5? « 




\3 


1 




5 


3 ^; 

> cq 


T 

00 
s 


S 

"33 
s: 

< 


2 

TO 
< 


S 

E 
a, 


a 
■« 

a, 
B 

TO 

O 


U 

S3 

c/5 


S 
S O 



^"3 



<3 

c 
« 

XI 
T3 
TO 
O 



it 

S 3 



§■ 



5 o « 



•1 E 

o (2 



<3 






§ 





tn 


'•"5 

s 

05, 


5 
3. 


3 
«j 
3 


V 




C) 


bo 


O 


^— ^ 


TO 


<J 




X> 
Xi 

TO 


3 


1 


U 


J 


u 



03 



I 
00 



I 
Oo 

S 



1 
-J 

? 



5 -s 



ex ^ 
« 00 

e := 
g s 

I » 
^ s 

E "O 
TO aj 

■C g 

1 g 



IS 



3 g 

S s g 

■§ -o -2 

I s s , 

'c 3 -p 

t-: ° _c 

5 ° -6 



ESS 

3 *-• £; 

« f 

£ <2 8 



j: °- 5 

OH- 
o -S wo 



E 



..I 



295 



296 



SALT-AFFECTED SOILS 



Table 11.4. Tolerance of various crops to percentage of exchangeable sodium in soils 3 



Tolerance to ESP and Range 




Growth Response Under 


at Which Affected 


Crop 


Field Conditions 


Extremely sensitive 


Deciduous fruits 


Sodium toxicity symptoms 


(ESP = 2-10) 


Nuts 

Citrus 

Avocado 


even at low ESP values 


Sensitive 


Beans 


Stunted growth at low ESP 


(ESP = 10-20) 




values even though the 
physical condition of the 
soil may be good 


Moderately tolerant 


Clover 


Stunted growth due to both 


(ESP = 20^0) 


Oats 


nutritional factors and 




Tall fescue 


adverse soil conditions 




Rice 






Dallisgrass 




Tolerant 


Wheat 


Stunted growth usually due to 


(ESP = 40-60) 


Cotton 


adverse physical conditions 




Alfalfa 


of soil 




Barley 






Tomatoes 






Beets 




Most tolerant 


Crested and Fairway 


Stunted growth usually due to 


(ESP = more than 60) 


wheatgrass 


adverse physical conditions 




Tall wheatgrass 


of soil 




Rhoades grass 





"From G. A. Pearson. 1960. U.S. Dept. Agric. Bull. 216. 
"ESP = exchangeable sodium percentage. 



nitely at a negative salt balance (more salt entering than leaving) with few adverse 
effects on soils or plants. The key in such cases is the amount of salt precipitating 
(and hence inactivated with respect to plants) in the soil. Normal plant growth can 
continue, provided that the quantities of salt precipitated do not lead to sodic soil 
conditions or to nutritional imbalances. 

The most common approach to salinity management is to maintain a prescribed 
leaching requirement (LR), defined as 



LR = 



^dw c*-iw 



th 



ECdv 



(11.10) 



where ECd W and EQ W are the electrical conductivities (salt concentrations) of the 
drainage and irrigation waters, and D m and Dd w are the amounts of irrigation and 
drainage water. The relationship is based on the assumptions that a salt balance exists 
(i.e., that ECj W Di W = EC(i W Dd W ) and that the plant is a perfect semipermeable mem- 



SALT BALANCE AND THE LEACHING REQUIREMENT 



297 



Table 11.5. Chloride tolerances of fruit varieties and rootstocks 3 

Limit to 
Tolerance to 
Chloride in 
Soil Satura- 
tion Extracts 
U 



Crop 



Rootstock or Variety 



(mmole L ' ) 



Rootstocks 



Citrus 


Rangpur lime, Cleopatra mandarin 


Citrus 


Rough lemon, tangelo, sour orange 


Citrus 


Sweet orange, cilrange 


Stone fruit 


Marianna 


Stone fruit 


Lovell, Shalil 


Stone fruit 


Yunnan 


Avocado 


West Indian 


Avocado 


Mexican 




Varieties 


Grape 


Thompson seedless, Perlette 


Grape 


Cardinal, Black Rose 


Berries 6 


Boysenberry 


Berries 


Olallie blackberry 


Berries 


Indian Summer raspberry 


Strawberry 


Lassen 


Strawberry 


Shasta 



25 

15 

10 

25 

10 

7 

8 

5 



25 

10 

10 

10 

5 

8 

5 



"From L. Bernstein. 1965. U.S. Department of Agriculture Information Bulletin 292. 



'Data available for a single variety of each crop only. 



brane removing only water from the soil solution and leaving all salts behind. The 
relationship is inaccurate when substantial salt precipitates in the plant root zone, 
dissolves from soil minerals, or is taken up by the crop. Despite these constraints, 
leaching requirement calculations are sufficiently accurate for most crop manage- 
ment purposes. 

Workers in Israel have demonstrated that careful management of extremely low 
leaching requirement values can still maintain adequate salinity control. One source 
of error in leaching requirement estimates is the substitution of EC of the saturation 
extract for the EC of the drainage water in the leaching requirement formula. Most 
salt-tolerance data apply to freely draining soil profiles. The salt concentration at 
saturation is commonly two to three times more dilute than in the water of a freely 
draining soil profile ("field capacity"). Hence, using saturation extract data in place 
of drainage water data gives an EC in the denominator of Eq. 11.10 that is 1/2 to 1/3 
too small, and a leaching requirement estimate that is 2 or 3 times too high. 

To calculate the leaching requirement one needs an estimate of the allowable EC 
of the saturation extract, such as can be obtained from existing salt tolerance data 



298 



SALT-AFFECTED SOILS 



(Table 1 1 .2). Soil texture or some other parameter must then be used to convert 
the EC value to an estimated EC of" the drainage water for the soil-plant system. 
This value and the EC of the irrigation water can be used to estimate the fraction of 
leaching water that must be passed through the plant root zone for salinity control. 
This excess water can then be compared to the soil infiltration rate, to plant tolerance 
at waterlogged conditions, and to drainage system capacity to see if salinity control 
is feasible under the chosen set of crop, soil, and water management conditions. 

The high leaching requirements that have been recommended in the past have 
taken into account nonuniform water distribution of many irrigation systems and the 
very great spatial variability of soil permeability. Excess water leaches through the 
more permeable parts of the field, while salts can remain in less permeable areas. 
Large quantities of water are therefore necessary to remove salts from the entire 
field. More uniform water application is possible with sprinkler irrigation or dead- 
level surface systems. More accurate estimates of the minimum leaching requirement 
can then be used. 



11.6 RECLAMATION 

The aims of reclamation are to make Ca 2+ the major exchangeable ion and to reduce 
the salt concentration in the soil solution. The main requirement to reclaim salt- 
affected soils is that sufficient water must pass through the plant root zone to lower 
the salt concentration to acceptable values. Passing 1 m of leaching water per meter 
of soil depth under ponded conditions normally removes approximately 80% of the 
soluble salt from soils (Fig. 1 1.6). If leaching is under unsaturated conditions, such 
as with the use of intermittent pending or sprinkler irrigation, this quantity of water 
may be reduced to as little as 350 to 200 mm or water. Boron removal can require up 
to three times more water than removal of Na and CI salts, because B is retained to 
some extent by soils. 



100 




5 1.0 1.5 2 2 5 30 3 5 4 4.5 
DEPTH OF LEACHING WATER PER 
UNIT DEPTH OF SOIL 

FIGURE 11.6. Depth of water per unit depth of soil required to leach a highly saline soil. (From 
R. C. Reeve et al. 1 955. Hilgardia 24:69-91 .) 



RECLAMATION 299 

Several techniques have been developed for reclaiming salt-affected soils. Pond- 
ing is a traditional method involving the construction of a large dike around the field. 
A substantial depth of water (commonly 0.3 m or more) is then maintained inside 
the dike to leach salts from the soil. Such an approach requires drainage facilities 
capable of removing large quantities of drainage water. The reclamation process is 
relatively inefficient, because much of the water passes through large soil pores that 
have already been purged of salts. Salt is removed only slowly from the fine poves of 
the adjacent soil mass. 

A more efficient leaching technique is the basin-furrow method. The soil is nearly 
leveled and irrigation water is allowed to meander back and forth across the field 
through adjacent sets of furrows. The water may take as long as a week to meander 
across the entire field under such conditions, but the quantities of water required are 
less than for ponded leaching. Furthermore, this technique does not produce sterile 
strips corresponding to former dike positions, where large quantities of salt accumu- 
late during ponded leaching. 

Soluble divalent ions (generally Ca) must be present during the reclamation of 
sodic soils. A common amendment for such purposes is gypsum (CaSC>4 • 2H2O), 
added at rates of up to several thousand kilograms per hectare in order to provide Ca 
as water percolates through the soil. Tests of soil exchangeable-sodium levels should 
be made every two to three years to estimate the need for reapplication of gypsum. 

Another Ca source is the lime in many salt-affected soils. If a soil is only slightly 
sodic and rather sandy in texture, tillage bringing subsoil lime to the surface before 
water application may be sufficient to maintain soil permeability during reclamation. 
Deep plowing to the 0.7- to 0.9-m depth has also proved helpful in redistributing sub- 
surface lime and in opening the soil to maintain adequate water permeability during 
reclamation. In most instances, however, lime is not sufficiently soluble to serve as an 
amendment for sodic-soil reclamation. It can be used as a source of soluble calcium 
only if an acidifying amendment is applied to dissolve the lime before reclamation 
begins. 

Common acidifying amendments for the reclamation of calcareous sodic soils are 
sulfuric acid and elemental sulfur. Sulfur must be oxidized to sulfuric acid by soil 
microorganisms before it becomes effective. A lead time of several weeks or months 
may be required for microbial oxidation before leaching begins. The reaction is 

2S + 30 2 + 2H 2 = 2H 2 S0 4 (11-11) 

CaC0 3 + H2SO4 = CaS04 + H 2 4- C0 2 (11.12) 

2NaX + CaS0 4 = CaX -+■ Na 2 S0 4 (11.13) 

Gypsum produced by the acid reacting with soil lime behaves like added gypsum 
during the remainder of the reclamation process. 

Still another reclamation procedure for sodic soils is the high-salt water recla- 
mation method. Saline-sodic soils will remain permeable as long as soil solution 
salt concentrations are high enough to flocculate the soil. The soil is leached with 
successively more dilute water while trying to increase exchangeable Ca by displac- 
ing exchangeable Na. Each increment of dilution is small enough to prevent the soil 



300 SALT-AFFECTED SOILS 

from swelling or dispersing. The initial step of adding high-salt water may increase 
either the soluble salt concentration or the exchangeable-sodium percentage of the 
soil. For example, treatments of soil with seawater, which has a salt concentration of 
600 mmol(-i-) L - ' and an SAR of approximately 60, generally produces soils with 
exchangeable-sodium percentages of 40 to 50. A fourfold dilution of the seawater, 
by mixing one part with three parts of freshwater, produces a salt concentration of 
150 mmol L~ ' and an SAR of 30 (because the SAR changes as the square root upon 
dilution of salt concentration). A second fourfold dilution step produces a salt con- 
centration of 37.5 mmol IT 1 , an EC of 3.8 dS trr 1 , and an SAR of 15. A third 
fourfold dilution produces an EC of 9 and an SAR of 7.5. This three-step leaching 
process would reclaim the soil if drainage facilities were adequate to remove the 
resultant large quantities of salty drainage water. The high-salt water reclamation 
method also depends on ready access to high-salt water. Saline ground and surface 
waters are common in salt-affected areas, and this is usually not a limitation. The 
ideal environment for high-salt water reclamation is near an ocean or saline lake, 
where both water access and disposal are available. 



APPENDIX 1 1 .1 THE LANGELIER INDEX 

Several workers have characterized the bicarbonate levels of waters with the Lange- 
lier Index (LI): 

LI = (pH a - pH c ) = pHa - (pK' 2 - p<.) + pC a + pAlk (11.14) 

where pH a is the measured pH of the soil or water, pH c is the calculated pH of the 
irrigation water if equilibrated with CaCC>3, pK' 2 is the second dissociation constant 
of H2CO3, pK' c is the solubility product of CaCO^, and pCa and pAlk are the negative 
logarithms of the molar Ca and molar(+) carbonate plus bicarbonate concentrations, 
respectively. The pH c can be derived from the reaction 

HCOJ = H + + CO^" (11.15) 

with its dissociation constant 

(H+)(CO; _ ) 

K 2 = — (11.16) 

(HCOp 

In a system at equilibrium with solid-phase CaC03, 

K c = (Ca 2+ )(COij-) (11.17) 

because the activity of solid-phase CaCC>3 can be taken as unity. Substituting 

Eq. 11.17 into 1 1.16 and rearranging gives 
> 

K 2 (HC07)(Ca 2+ ) 
(H+)=— -^ - (11.18) 



QUESTIONS AND PROBLEMS 301 

or, in terms of the concentrations of HCOJ and Ca 2+ , 

(H + ) = ^C0 3 [HC0 3 -Iyc a [Ca^ (iu9) 

where brackets indicate concentrations and the y's are activity coefficients. Taking 
negative logarithms gives 

pH c = vK' 2 - pK' c + p(yHC03) + p(yCa) + pfPCO^l + p[Ca] (l 1.20) 

Combining the two activity coefficients with the dissociation and solubility constants 
gives 

P H = (p^ - V K' C ) + p[HC0 3 -] + p[Ca 2+ ] (l 1 .21) 

where the pA" terms are treated as a single, concentration-dependent quantity (such 
"constants" are truly constant only if expressed in terms of activities). Except at pH 
>9, when appreciable carbonate exists, Eq. 1 1.21 is equivalent to the pH^ portion of 
Eq. 11.14. 

The fraction of bicarbonate precipitating from an irrigation water is often related 
linearly to both pH c and the leaching fraction. With the advent of computers, how- 
ever, it is more common now to predict the amount of CaC03 precipitating under a 
given set of management conditions from irrigation- water chemistry, initial and time- 
course soil chemistry, the leaching fraction, and measured or inferred CCh levels of 
the soil atmosphere. 



BIBLIOGRAPHY 

Ayers, R. S., and D. W. Westcot 1976. Water Quality for Agriculture. Food and Agriculture 
Organization of the UN, Irrigation and Drainage Paper 29, Rome. 

Bernstein, L. 1964. Salt Tolerance of Plants. United States Department of Agriculture. Infor- 
mation Bulletin 283. 

United States Salinity Laboratory Staff. 1 954. Diagnosis and Improvement of Saline and Alkali 
Soils. United States Department of Agriculture. Handbook 60. 



QUESTIONS AND PROBLEMS 

1. An arid area receives 150 mm of rainfall annually with an average salt concen- 
tration of 10 mg L~ l . 11' the surface soil from this area contains an average of 
30% water at saturation, how many years would be required for sufficient salt to 
be added from the atmosphere to increase the EC of the saturation extract by 1 
dSm-'? 

2. An irrigation water contains 750 mg L _l soluble salts. If used at an average 
leaching fraction of 0.15, what would be the average EC of the drainage water 
leaving the bottom of the crop root zone? 



302 SALT-AFFECTED SOILS 

5. What is the EC (in dS m~ } ) of a solution having an electrical resistance of L500 
Q in a conductivity cell with a cell constant of 5.0 cm -1 ? 

4. An irrigation water has an EC of 0.8 dS m _l and a sodium concentration of 35 
mgL -1 . Calculate: 

(a) Its osmotic potential. 

(b) ItsSAR. 

(c) The equilibrium ESP for soils having a Gapon exchange constant of 0.015 
(Lmmor 1 )'/ 2 . 

5. An irrigation water contains 7 mmol(-f) L _1 total cations, 1.5 mmol(+) L -! 
Ca 2+ , 1 mmoJ(+) L -1 Mg 2+ , and 5 mmoi(+) L _l HC(\. Calculate the pH c 
and the adjusted SAR of this water. To what extent would the water be regarded 
as hazardous? 

6. If the irrigation water of Problem 5 has a pH of 7.8, what is the residual sodium 
carbonate value? To what extent would the water be regarded as hazardous based 
on this criterion? 

7. An irrigation water contains 600 mg L~' TDS and has an SAR of 6. If it is 
applied to a soil containing 40% water at saturation and 20% water at "field 
capacity," what will be the EC and SAR of the saturation extract of surface soil 
and the resultant salinity classification category after prolonged irrigation with 
this water? if the same water-holding characteristics are found at the bottom of 
the crop root zone, what will be the EC and SAR of the saturation extract for soil 
from this portion of the profile after prolonged irrigation at a leaching fraction 
of 20%? 

8. For a soil having the water-holding characteristics described in Problem 7, what 
will be the leaching requirement if an irrigation water of EC = 0.8 dS m - ' is 
used to irrigate alfalfa under conditions where no more than a 25% yield reduc- 
tion due to salinity can be tolerated? 

9. A 30-cm depth of surface soil contains 28% exchangeable sodium and has a CEC 
of 150 mmol(-l-) kg -1 . How many tonnes per hektar of sulfur will be required 
to lower its ESP to 5%? 

10. Explain how overgrazing, conversion from shrubs to grasses, burning, and sum- 
mer fallowing can lead to "saline seeps." 

11. Explain why soils containing appreciable amounts of zeolites may give unreli- 
able ESR/SAR relations, and why the SAR is the more accurate index in such 
soils. 

12. Explain how the salt tolerance data of Table 1 1 .2 might be related to "real-world" 
salinity distributions, such as shown in Figs. 1 1.2, 1 1.3, and 1 1.5. 



INDEX 



Acid: 

Bronsted, 71 

dissociation constant, 71 

Lewis, 84 

ram, 61, 187 

soil, 260 

soil management. 274 
Acid sulfate soil, 1 16, 260 
Acidity, exchangeable, 266 

fertilizer. 260 

measurement, 268 

reserve, 40 

soil, 187,260 

(itratablc, 265 

weathering progression, 176 
Activation energy, 98 
Activity, chemical, 76, 275 

ion product 86, 94 

solid, 96, 275 
Adsorption: 

edge, 222 

Freundlieh, 254 

gas, 253 

hydrophobic, 252 

isotherm (equation), 254 

Langmuir, 254 

multi-site. 256 

negative, 242 

ovgunochemical, 170 

site, 223 

specific & nonspecific, 243 

weak acid, 245 
Aeration, soil, 108 
Alkali cation, 35 
Alkaline earth cation, 35 
Alkalinity, 286 



Allophane, 130, 144, 197 
Aluminium, 32, 50, 145, 263 

hydroxyoxide, 145, 173 
Ammonia, 7 1 , 260 
Amorphous, 144 
Amphoterism, 75, 274 
Anaerobic, 108 
Analcime, 145 
Anatase, 146 
Anion, soluble, 33 

exchange (retention), 237 

exchange capacity. 134, 240 

penetration, 244 

repulsion, 2 1 7 
Antigorite, 140 
Apatite, 49 
Aquept, 156 
Area, global soil, 186 

salt-affected soil, 280 
Aridisol, 156 
Arsenic, 46 
Atmosphere, 14, 17 
Atomic, number, 19 

weight, 19 
Augite, 18! 

Avogadro's number. 19 
Azotobaclec, 62, 167 

Base, 71 

Lewis, 84 

saturation, 268 
Bentonitc, 137 
Beryllium, 30 

BET (Brunacr-Emmeu-Tellcr), 141, 254 
Bicarbonate, 300 

hazard, 288 



303 



304 



INDEX 



Biogcochcmical cycle, 1 1 
Biomass, microbial, 155 
Biosphere, J 7 
Biotite, 137, 143 
Blue-green algae, 62 
Boehmite, 145 
Bonding: 

covalcnt, 131 

electrostatic, 130 

hydrogen, 1 3 1 , 25 1 

ionic, 130 

van (icr Waals, 131,25! 
Borate, boron, 45, 237, 289 

plant tolerance, 295 
Bragg'slaw, 152 
Breakthrough curve, 249 
Bromide, 42 

Bronsted acid and base, 7 1 
Brucile. 138 
Bulk solution, 208, 21 7 

Cadmium, 30, 53, 224 

Calcite, 130 

Calcium, 37 

Carbohydrate, 91, 109 

Carbon, see also Soil organic matter, 61 

cycle, 14 
Carbonate, 197 
Carbon dioxide, 198 
Catalyse, 98 
Cat clay, 260 
Cation: 

exchange (retention), 206 

capacity, 134, 146, 167, 206, 229, 240 
divalent, trivalent, 52, 221 
equation, 215, 229 
selectivity, 212 

exchangeable, see also Lyotropic series, 35 
Charge, permanent, 146 

pH-dcpcndent, 147, 264 
Chelate, 81 
Chert, 204 
Chloride, 42, 237 

plant tolerance. 294, 297 
Chlorite. 138. 143 
Chromium, 53 
Clay mineral, 129, 135 

composition. 180 
Climate and soil, 176 
C/N/S ratio, 167 
Cobalt, 33, 53, 222 
Colligativc property. 76 
Colloid, 7 
Complementary cation. 21 1 



Complex ion, 79 
Composition, plant and soil, 8 
Conductance, electrical, 285 

water, 286 
Consumption, 12 
Coordination number, 1 32 
Corundum, 145 
Cryplocrystallinc, 144 
Crystal, 131 

radii, 134 

Duvies equation, 78 
Davis equation, 230 
Debye-Hueckel equation. 77 
Deduction, 5 
Deficiency, 30 
Dcnitrincation, 62 
Density, earth, 26 
Deprotonation, 72 
Dialysis, 261 
Dioctohcdral, 138 
Dissolution, congruent, 174 
Distance of closest approach, 78 
Dolomite, 38 
Donnan equation, 232 
Double layer, diffuse, 3, 217, 225 
thickness, 219 

EDTA, 81 

Eh, 117 

Electrical conductivity (EC), 285 

Electrochemistry, 116 

Electrode, ion-selective, 107, 275 

pH, 275 

platinum, 1 14 

potential, 1 1 3 

reference, 276 
Electron, activity and availability, 1 17 

acceptor, 108 

donor, 108 
Electron reversibility, 1 26 
Electrostatic repulsion, attraction, 217 
Elovich equation, 100 
Enthalpy, 89 
Entropy, 88 
Enzyme kinetics, 97 
Equilibrium, 88 

constant, 80 
Escaping tendency, 94 
Essential element, 3, 27 
Evaporite, 200 

Exchangeable sodium ration (ESR), 287 
Exchange capacity. See cation and anion 
exchange 



INDEX 



305 



Exclusion volume. 242 
Expanding clay, 137 

Feldspar, 129, 181 

Fermentation, 1 1 2 

Fertilizer, See element in question 

Held capacity, water, 297 

Flooded soil, 1 15 

Fluorine, 245 

Food production, 4 

Free energy, Gibbs, 90 

Ful vie acid, 162 

Gas, diffusion, 108 

ideal, 23 
Gapon equation, 216 
Gibbsite, 50, 145 
Global, carbon/nitrogen/sulfur, 61 

water, 13 
Goethite, 145 
Gypsum, 130 

Half-life, 101,160 
Half-reaction, 109 
Halloysitc, 140 
Halophyte, 40 
Hardness, water, 286 
Heavy metal, 53 
Hclmholtz double layer, 2 1 7 
Hematite, 145 
Henry's law, 199 
History, 5 
Histosol, 116, 156 
Humicacid, 162 
Hum'm, 162 
Humus, 155, 163 
Hydration, heat of, 70 

number, 69 
Hydrogen, 50. See also pH 
Hydrogen clay, 260 
Hydrolysis, 72 
Hydroniuin ion, 71 
Hydrosphere, 12, 1 8 
Hydrous mica, 1 43 
Hydroxyoxtde. 143. 73 
Hysteresis, 210 

Igneous mineral, stability, 181 

weathering rate, 182 
Illite, 143 
Imogolite, 144 
Inductive reasoning. 5 
Inhibitor, reaction,, 98 
Iodine, 42 



Ion, activity product. 86, 94 

packing, 131 

radii, 184,212 
Ionic potential, I S3 
Ionic strength, 77 
Ion pair, 79 
Iron, 32, 145 

redox stability, 124 
Isomorphic substitution, 134, 182 
Isotope, 20 

Junction potential. 268, 276 

Kaolinite, 136 
Kerr equation, 2 1 5 
Kinetics, 'H 
K.OC Kqw. 252 

Langelier index, 300 

Lanthanide, 52 

Laterite, U, 146 

Layer silicate, 135 

Leaching fraction, requirement, 294 

Lead, 30, 53, 224 

Lewis acid and base, 84, 224 

Life, origin of, 15 

Ligand, 72, 79 

exchange, 244 
Lignin. 110 
Lime requirement, 269 
Lyotropic series, 2 1 3 

Macroelement, -nutrient, 30 
Mafic, 144,281 
Maghemite, 146 
Magnesium, 38 
Magnetite, 146 
Manganese, 145 
Maturity, soil, 176, 189 
Mechanism, 97 
Mercury, 30, 53 
Mesophile, 102 
Metamoiphic rock, 181 
Mica, 137, 143, 181 
Mtctiuelis-Menten, 103 
Microelement, -nutrient, 30 
Mineral, accessory, 144 

formation, 192 

identification, 152 

intcrgrade, 140 

primary, 129, 193 

secondary, 129, 177 

soil-formed, 190 

weathering, 188 



306 



INDEX 



Mineralization, 62 
Mixing, homogeneous, 94 
Molarity, 20 
Mole, 19 

fraction, 2 1 

of ion chargc,20 
Molecular retention, 240, 250 
Molecular sieve, 1 45 
Mollisol, 155 
Molybdenum, 33, 46, 273 
Montmorillonite, 137, 141 
Muscovite, 137, 143 

Ncrnst equation, 1 17 
Nitrate, 4) 
Nitrogen, 61, 167 

fixation, 62 

redox stability diagram, 1 1 8 
Nutritional imbalance, 294 

Osmotic, adjustment, 291 

potential, 286 
Overvollage, redox, 1 16 
Oxidation, 33, 107 
Oxide, free, 130 
Oxidizing agent, 108 
Oxisol, 184 
Oxyanion, 44 
Oxygen, soil, 108 

Parent material, 173, 190 

Partial pressure, 23 

Partitioning, octanol-water, 252 

pe, 1 17 

Peal. 157 

Pedogenic, 178 

Periodic Table of die Elements, 2 1 , 28 

Pesticide, adsorption 240, 250 

pH,20, 107 

crop yield, 270 

essential elements, 272 

flooded soil, 108 

measurement, 268, 275 

soil, 40, 175 
pH c ,301 
Phosphate, 33, 47, 179, 245 

fertilizer, 49, 66 

stability diagram, 247 
Phyllosilicatc, 136 
Phytolith, 45, 197 
Plagioclase, 181 
Plant, acidity tolerance, 270 

composition, 8 

salt tolerance, 291 



Plinthite, 11, 146 
Polarity, 68 
Pollution, 1 1 , 27 
Potassium (kalium), 38 

availability, 273 

fixed, 143, 214 
Potential, chemical, 76 

diffusion, 268 

electrode, 112 

irreversible, 125 

liquid-liquid junction, 268 

osmotic, 227 

platinum, 125 

redox, 125 

surface, 227 

water, 76, 286 
Productivity, net primary, 158 
p scale, 20 
Psychrophile, 102 
Pyrolusitc, 146 
Pyrophillite, 138 
Pyroxene, 181 

QlO. 102 . 
Quartz, 129, 204 

Radioactive element, 31 
Radium, 35 
Rate, constant, 99 

reaction, 97 

temperature effect, 101 
Reaction order, 99 
Redox, 107 

couple, 1 1 3 
Reduction, 33, 107 
Residence time, 1 86 
Residual sodium carbonate, 288 
Rhizobium, 62 
Rock, composition, 173, 178 

sedimentary, 181 
Rutile, 146 

Saline soil, 280 

characterization, 289 

effect on plants, 291 

reclamation, 298 

seep, 282 
Salt: 

balance, 294 

bridge, 275 

fossil, 281 

tolerance, plant, 291 
Salting coefficient, 79 
Saltpan, 282 



INDEX 



Sand, 129 
Saturation index, 97 
Scientific method, 5 
Selenium, 32, 44 
Serpentine, 38, 181 
Sesquioxide, 179 
Silica, soluble, 204, 241 
Silicate, 135 

structure, 130 
Silicon, 44 
Silt. 129 

Smectite, 137. 141 
Sodicily, 286 
Sodium (natrium), 39 

hazard, 286 

plant tolerance, 296 
Sodium adsorption ratio (SAR), 287 
Soil: 

acidity, 50, 187,260 
classification, 265 

composition, 8. 178 

development, 172 

and environment, 2, 17 

and life, 1 5 

organic matter, 108 

saline. 280 

sodic,29l 

toxicity, 26 

water, 13 

weathering rate, 185 
Soil-forming factor, 156, 185 
Soil organic matter, 108, 155 

colloidal property, 164 

composition, 163 

decay, 158 

extraction, 161 

function, 167 

half-life, 160 
Soil solution, I 

model, 87 
Sol, 7 
Solid, activity, 48 

inorganic, 129 

organic, 155 
Solubility product, 73 
Solution, aqueous, 68 

ideal aqueous, 76 

solid, 68, 93, 222 
Solvation sheath, sphere, 69 
Spontaneous reaction, 91 
Stability diagram, Eh, 1 18 

mineral, 201 
State, of a system, 88 

standard, 91 



Stern layer. 228 
Stoichiometry, 209 
Strengite, 49 
Sulfide, 116 
Sulfur, 42, 65, 1 67 

redox diagram, 1 22 
Surface area, 8, 151, 186 

external and internal, 138 
Symmetry, 212 
System, closed, 88 

open, 93 
SI (Systeme International d' Unites), 22 

Talc, 138 

Tetrahedva linkage, 182 

Thermodynamics, 87 

exchange constants, 230 

irreversible, 89 
Thermophile, 102 
Titanium, 52, 145, 191 
Total dissolved solids (TDS), 284 
Toxic, ion, 27, 30, 53 

aluminium and manganese, 272 

sodium and chloride, 294 

solute, 289 
Trace element and metal, 53 
Transition clement, metal, 33, 50 
Trioctahedral, 138 
Tropical soil, 178, 240 
Turnover rate, 186 

Unit cell, 134 

Valence dilution, 21 1 

Van der Waals bond, 131,251 

Vanselow-Argersinger equation, 229 

Variable charge, 240 

Variscite, 49 

Vermiculite, 137, 142 

Vertisol, 181 

Water, 68 

field capacity, 297 

global, 13 

irrigation water quality, 284 

natural, composition, 203 

redox stability, 1 19 
Weathering, agent, 1 72 

losses, 186 

reversed, 201 

Xray diffraction, 152 
Zeolite. 145,200